This Week In Startups

In the episode "Ask Jason LIVE! Navigating startup growth with Real-Time Q&A E1937" of the podcast "This Week in Startups," Jason Calacanis engages with various entrepreneurs seeking advice on their startup challenges. The episode features a series of questions from startup founders and entrepreneurs, each presenting unique scenarios and seeking strategic guidance from Jason, who draws on his extensive experience in the startup ecosystem. One of the guests, Dustin, discusses his transition from shutting down his startup to seeking roles in established companies to gain stability and learn from successful founders. Jason advises him on positioning himself as an operations specialist and leveraging his generalist skill set to appeal to companies looking for versatile problem solvers. He emphasizes the importance of demonstrating the ability to handle various tasks effectively, which is highly valued in startups. Another significant discussion involves a founder who operates a platform offering premium experiences and revenue models for independent brick-and-mortar businesses. The service allows businesses to offer memberships or subscriptions, enhancing customer loyalty and revenue. Jason probes into the business model, customer acquisition, and scalability potential. He advises focusing on proving the concept through metrics like customer satisfaction and churn rate, suggesting that the venture might not immediately attract traditional VC interest but has potential due to its innovative approach to enhancing small business revenues. Throughout the episode, Jason provides tailored advice, focusing on practical steps the entrepreneurs can take to refine their business strategies, enhance their market positioning, and address operational challenges. His guidance often revolves around understanding customer needs, leveraging unique business strengths, and strategically using resources to foster growth and stability in their ventures.

In the podcast episode "AI Demos and News Llama 3, Marblism, Lumona & AI Jet-Fighter Dogfights E1936," the hosts discuss various topics related to artificial intelligence and its applications. They start by examining Facebook's recent advancements in AI, particularly their development of Llama 3, a language model that has shown impressive capabilities despite its smaller size compared to other models like GPT-4. The hosts express surprise at Facebook's rapid progress in AI, noting that Llama 3 is not only more efficient but also free and open source, which could have significant implications for businesses and developers. The conversation then shifts to the broader implications of open-source models versus commercial AI products. The hosts debate the potential benefits and drawbacks of using open-source AI, such as increased accessibility and customization, against the support and reliability that might come with commercial AI solutions from companies like OpenAI or Salesforce. They also touch upon the strategic reasons companies like Facebook might choose to open-source their innovations, suggesting that it can lead to broader industry improvements and indirectly benefit the original innovator. The episode also covers practical applications of AI in various fields. The hosts discuss an AI-driven dogfight involving an F-16, highlighting the potential of AI to outperform humans in complex tasks and its implications for military training and operations. Additionally, they explore new AI tools and platforms that are making it easier for developers and businesses to implement AI solutions. For example, they mention Marblism, a tool that allows users to create fully functional apps from simple prompts, and Lumona, a search engine that aggregates product information from multiple sources to provide comprehensive answers to user queries. Throughout the episode, the hosts maintain a critical perspective on the trustworthiness of big tech companies in handling AI advancements, particularly Facebook. They recount anecdotes of Facebook's past behavior with partners and developers, using these to frame their cautious optimism about the company's current direction in AI. The episode concludes with a discussion on the potential of AI to transform various industries and the importance of ethical considerations and transparency as AI technologies become increasingly integrated into everyday life.

In a recent episode of "This Week in Startups," the focus was on Meta's significant strides in AI, particularly with the release of Llama 3, their latest open-source model. The discussion highlighted Meta's transformation from a company primarily focused on social media and the metaverse to a formidable player in the AI and search engine space. The rebranding to Meta.ai was seen as a strategic move to redirect the company's focus towards leveraging AI technology, particularly in enhancing search functionalities across its platforms. The episode delved into the technical aspects of Llama 3, explaining its efficiency and how it compares to other AI models like GPT-4. With 70 billion parameters, Llama 3 is positioned just behind GPT-4 in performance but is significantly more efficient due to its smaller size. This efficiency could potentially revolutionize how AI models are developed and deployed, emphasizing the importance of quality data over the sheer size of the model. Furthermore, the discussion touched on Meta's strategic use of open-source models to potentially democratize AI technology. By making Llama 3 open-source, Meta not only fosters a collaborative environment that could accelerate improvements and innovations in AI but also strategically positions itself against other tech giants by potentially reducing the costs associated with AI development. The podcast also explored the broader implications of Meta's AI advancements on the tech industry, particularly in terms of competition with Google. Meta's integration of AI-driven search functionalities across its platforms could significantly impact Google's dominance in the search engine market. This move by Meta, to embed advanced AI capabilities directly into its widely used platforms, could shift the dynamics of online advertising and search, presenting new challenges and opportunities for all players in the tech ecosystem. Overall, the episode painted a picture of a rapidly evolving tech landscape where AI is becoming a critical focal point for innovation and competition. Meta's aggressive push into AI with Llama 3 and its broader strategic implications could potentially reshape how businesses and consumers interact with technology on a daily basis.

In the podcast episode, Logan Bartlett from Redpoint Ventures and Hunter Somerville from StepStone Group join Jason Calacanis to discuss various topics including venture capital deployment, mergers and acquisitions, and Google's strategy on generative AI. The conversation begins with a discussion on the current state of fundraising in the venture capital space, highlighting the challenges faced by late-stage funds and the success of established platform brands in securing capital. Hunter Somerville notes that big allocators are inclined towards well-known brands for safety and substantial investments, while emerging managers struggle in a competitive fundraising environment. The discussion then shifts to the strategy of taking partial exits through secondary sales, which has become a prudent approach due to the elongated hold periods in the industry. This strategy helps in managing fund liquidity and maintaining investor confidence, especially when full exits are delayed. The conversation also covers the impact of Tiger Global's investment approach, which was characterized by rapid deployment of capital with minimal due diligence. This approach led to discussions on the importance of governance and the need for venture capital firms to balance aggressive growth with responsible management to ensure long-term sustainability. On the topic of Google's AI strategy, the panel discusses the tech giant's significant investment in generative AI, noting the competitive landscape with other major players like OpenAI and Microsoft. The conversation highlights the rapid advancements in AI technologies and the potential implications for various industries, including venture capital. Throughout the episode, the speakers emphasize the importance of strategic fund management, the challenges of maintaining high standards in a fluctuating market, and the critical role of innovation in sustaining growth and competitiveness in the venture capital industry.

In the podcast episode titled "Destiny (DXYZ) and the Buzz Around Public Access to Private Companies," the discussion revolves around the challenges and motivations behind creating a publicly traded fund that allows average investors access to private tech companies. The guest, Sohal Prasad, explains the inception of Destiny Tech 100, a fund designed to make private tech investments accessible to the general public through a brokerage account. This initiative was driven by Prasad's personal desire to offer investment opportunities in private tech to people like his father and friends, who otherwise wouldn't meet the stringent criteria set for accredited investors. Prasad details the structure of Destiny Tech 100 as a listed closed-end fund, which differs from traditional ETFs and other investment vehicles like SPACs and interval funds. He highlights the regulatory challenges and the extensive process involved in setting up such a fund, including navigating SEC regulations and working without traditional financial institutions as underwriters. The fund, which started with contributions from over 200 individual investors, primarily successful tech entrepreneurs and VCs, aims to democratize investments in private companies by making shares available on public exchanges like the NYSE. The episode also touches on the broader implications of making private company investments available to the public. Prasad discusses the potential market volatility and the public's reaction to the fund's performance, emphasizing the long-term vision behind Destiny Tech 100. He addresses the complexities of managing a fund that deals with private company shares, including dealing with company founders and the secondary market dynamics. Overall, the podcast sheds light on the innovative approach of Destiny Tech 100 in bridging the gap between private tech investments and the average investor, aiming to transform how people can participate in the growth of private companies before they go public.

In episode 1932 of "This Week in Startups," hosted by Jason Calacanis, a startup pitch competition takes place where three founders present their businesses in hopes of securing a $25,000 investment from Jason's venture capital fund, Launch. The episode features discussions on the importance of having co-founders and the ability to build a product as critical factors for entrepreneurial success. The podcast also highlights the Founder University program, a free accelerator designed to help nascent startups navigate the early stages of business development. The first founder to pitch is Jared from Umbrella Sports, who introduces SmartGrip, a hardware device designed to improve golfers' grip and swing. The product has already seen success through a crowdfunding campaign and is complemented by a companion app that offers additional features and subscription-based content. Jared outlines ambitious revenue goals and discusses the competitive pricing and market potential of SmartGrip in both consumer and business markets. Next, Linda from Mastertech AI pitches her platform aimed at revolutionizing auto repair by using AI to aggregate repair data and guide technicians through repairs with an AI conversational interface. The platform intends to license OEM data to ensure accuracy and plans to charge auto repair shops a monthly subscription fee. Linda's pitch emphasizes the extensive market opportunity and her team's deep expertise in both technology and the auto repair industry. The third pitch comes from Alexander of Ellis, who presents a unified B2B email marketing platform that automates and personalizes email campaigns to improve engagement and conversion rates. Ellis aims to simplify the complex process of setting up and running email campaigns by integrating various services into one platform. Alexander discusses the early traction of the platform and its pricing model, which includes a monthly subscription and a fee per contact. After hearing all pitches, Jason deliberates on the potential of each startup, considering factors like market size, product innovation, and the founders' ability to execute. The episode concludes with Jason deciding to invest in all three companies, highlighting the diverse opportunities and promising prospects each founder presents.

In the podcast episode titled "Udio AI’s Music Magic, Aqua’s Unique Dictation, Copyright Laws & the 1000x Potential of AI," the hosts discuss the transformative potential of AI in various industries, particularly in software development. They highlight how AI can significantly increase productivity, allowing software engineers to tackle multiple problems simultaneously with the help of AI agents. This could lead to a 1000x increase in software development capabilities, revolutionizing the industry. The episode also covers the launch of new AI chips by major companies like Meta, Google, and Intel, which are set to enhance the AI ecosystem by providing more efficient hardware solutions. This development is expected to drive down costs and increase competition, benefiting the industry overall. Additionally, the podcast delves into copyright issues related to AI, particularly focusing on a new bill that would require disclosure of the sources of content used by AI systems. This segment discusses the ethical and legal implications of using copyrighted material without permission to train AI models, which has been a controversial practice in the industry. The hosts also introduce Udio, a new service backed by tech and music heavyweights that uses AI to create music. They discuss the potential of AI in creating music that could rival human compositions, although they also touch on the legal complexities involved, especially regarding the use of copyrighted material without proper licensing. Lastly, the episode explores Aqua, a unique dictation tool that enhances note-taking by allowing users to edit and format text through voice commands. This tool represents another facet of AI's integration into daily tasks, making technology more accessible and efficient.

In the podcast episode, Tomasz Tunguz and David Clark discuss the intricacies of investing in AI and the startup valuations for Q1 2024. The conversation begins with an overview of the recent Y Combinator demo day, highlighting the dominance of AI startups and the significant interest they have garnered from venture capitalists. Tomasz Tunguz of Theory Ventures elaborates on the opportunities within AI, distinguishing between co-pilots and agents, and their respective impacts on productivity and job replacement. He emphasizes the strategic necessity for startups to focus on vertical AI to effectively compete against tech giants like Microsoft and Google. David Clark from Vencap shares insights on the investment landscape, particularly how startups are adapting to the competitive environment by focusing on specialized applications of AI. The discussion also touches on the challenges and strategies related to managing venture funds and making investment decisions in a rapidly evolving tech landscape. The podcast also delves into the broader implications of current market trends on startup valuations and venture capital. With insights from the recent data released by Carta, the conversation covers the complexities of navigating an investment environment where entry prices are rising despite potentially depressed exit valuations due to a sluggish IPO market and regulatory constraints on large M&A activities. The dialogue concludes with a discussion on the importance of strategic reserve management in venture funds and how it can significantly influence the success of investments in high-growth startups.

In episode 1929 of "This Week in Startups," host Jason Calacanis discusses the future of flying cars with Jim Dukhovny, CEO of Alef. Alef is pioneering the Model A, a vehicle that represents a significant leap towards the realization of flying cars, a concept that has captivated imaginations for over a century. Unlike traditional cars or even the vertical takeoff and landing (VTOL) vehicles that have been in development for years, the Alef Model A offers a unique approach to achieving flight by essentially turning the car on its side, using the vehicle's sides as wings. This design eliminates the need for large, retractable wings, making the vehicle lighter and more practical for vertical takeoff and landing. Dukhovny explains that the technology for autonomous flight in the Alef Model A already exists, including the necessary software and hardware. However, the path to certification for autonomous human flight is currently non-existent, posing a significant barrier to implementation. Despite this, Alef has successfully conducted flights with full-scale models, proving the concept's viability. The vehicle is designed to be incredibly light, enhancing its safety features, such as the ability to deploy ballistic parachutes at low altitudes. Alef's approach to flying cars is not just about overcoming technical challenges but also about navigating regulatory landscapes and societal acceptance. The company has opened pre-sales for the Model A, with over 3,000 pre-orders, indicating significant interest in the technology. Dukhovny is cautious about providing a specific timeline for widespread availability but suggests that the first models could start production by the end of 2025, assuming all goes well with funding, regulations, and technological development. The conversation also touches on the broader implications of flying cars, including the potential for safer transportation compared to traditional cars, given the reduced likelihood of collisions and obstacles in the air. Dukhovny envisions a future where flying cars are not only a reality but are also accessible and affordable to the general public, transforming urban mobility and potentially making the dream of flying cars a practical solution for everyday transportation.

In episode E1928 of "This Week in Startups," host Jason Calacanis interviews Ramin Hasani, CEO and co-founder of Liquid AI, about the innovative approach his company is taking towards artificial intelligence (AI) through the development of liquid neural networks. Hasani explains that the mission of Liquid AI is to design AI systems from first principles, rooted in biology and physics, leading to the creation of liquid neural networks during his PhD program. These networks are inspired by the nervous system of the C. elegans worm, which shares 75% of its genes with humans and has a fully mapped nervous system. Unlike traditional AI systems that are fixed after training, liquid neural networks remain adaptable, offering more dynamic and robust responses. Hasani's work focuses on making AI models smaller, more efficient, and capable of running on minimal hardware like a Raspberry Pi, without sacrificing performance. This approach could revolutionize the AI field by reducing the energy and computational resources needed for training and deploying AI models. Liquid AI has raised significant funding and is collaborating with major system integrators worldwide to commercialize their technology across various sectors, including finance, healthcare, and automotive. The conversation also touches on the broader implications of AI advancements, including the race towards Artificial General Intelligence (AGI), the potential for AI to solve major global challenges, and the ethical considerations of AI development and deployment. Hasani believes that within the next two to five years, significant leaps in AI capabilities will be witnessed, possibly leading to the first versions of AGI. He emphasizes the importance of explainability in AI systems to ensure they are understandable and controllable, contrasting Liquid AI's approach with the black-box nature of larger, less interpretable models currently dominating the industry. Overall, the episode provides a deep dive into the cutting-edge work being done at Liquid AI, the potential of liquid neural networks to transform AI, and the philosophical and practical considerations of advancing towards AGI.

In this episode of the Liquidity Podcast, hosted by David Weisberg with guests Apoorva Mehta from Summit Peak, Jack Altman from Alt Capital, and Jason Calacanis from The Launch Fund, the discussion revolves around the evolving dynamics of venture capital, the role of boards in startups, and the impact of liquidation preferences on startup valuations. Jack Altman shares insights on the value of having experienced, invested board members in guiding startups through their journey, emphasizing the long-term relationship and outside perspective they bring, which is often more consistent than that of any executive team member. He also discusses his dual role as an investor and operator, highlighting the challenges and learning curves associated with building a company and how this experience informs his investment decisions. Apoorva Mehta talks about the criteria for investing in operator VCs and the balance between operating a company and managing a venture fund. He stresses the importance of fund size matching the portfolio strategy and the general partner's ability to source, pick, and win deals. Mehta also touches on the trend of solo GPs and the discipline observed across the venture capital industry in terms of portfolio construction and valuation marks. Jason Calacanis reflects on the strategic advantage of disciplined portfolio management and the importance of governance in startups. He shares his approach to board participation, aiming to provide founders with guidance based on his experiences. Calacanis also discusses the significance of acquisitions in the startup ecosystem and expresses concerns about regulatory challenges that could stifle M&A activity. The conversation also delves into the resurgence of premium liquidation preferences in venture capital deals, with the guests discussing how these terms can affect startup valuations and the potential risks for employees unaware of the financial structures that could impact their equity compensation. Throughout the episode, the guests share their latest investments, highlighting a focus on AI and enterprise software startups. They discuss the potential of AI to revolutionize various sectors and the importance of supporting early-stage companies in this rapidly evolving landscape. Overall, the episode provides valuable insights into the venture capital ecosystem, the strategic role of boards in startups, and the evolving landscape of startup financing and valuation.

In the podcast episode titled "Growth Hacks for Startups from Elliot Shmukler of InstaCart, LinkedIn and now Anomalo E1926," Elliot Shmukler shares his extensive experience and insights into startup growth strategies. Shmukler, who has played significant roles in the growth of companies like Instacart, LinkedIn, and Wealthfront, emphasizes the importance of word of mouth as a foundational growth channel for startups. He argues that no great product exists without phenomenal word of mouth, highlighting that people telling their friends about a product is the most effective growth strategy. Shmukler discusses the concept of product-market fit and its critical role in a startup's growth phase. He advises that before achieving product-market fit, startups should focus on acquiring initial users through any means necessary, including non-scalable methods. Once product-market fit is established, startups can then systematically think about growth. He also touches on the significance of having a growth mindset across the company and the potential roles within a team dedicated to growth. The conversation also covers various growth strategies and tactics, including building virality into the product, creating products that people want to brag about, and implementing referral programs. Shmukler shares anecdotes from his time at Instacart and Wealthfront, illustrating how specific strategies led to significant growth. He also discusses the challenges of managing growth, such as dealing with duplicate accounts and the importance of data quality in making informed growth decisions. Shmukler introduces his current venture, Anomalo, which aims to address data quality issues that companies face. Anomalo provides a tool that uses AI and machine learning to automatically detect and alert companies about data issues, helping prevent the negative consequences of bad data on growth efforts. He explains how Anomalo fits into the broader landscape of data management and its importance for companies heavily reliant on data for decision-making. Overall, the podcast episode offers valuable insights into growth strategies for startups, emphasizing the importance of word of mouth, product-market fit, and data quality. Shmukler's experiences and the examples he shares provide practical advice for startups looking to accelerate their growth.

In the podcast episode titled "CoreWeave’s Brannin McBee on the future of AI infrastructure, GPU economics, & data centers," Brannin McBee, the Chief Development Officer and co-founder of CoreWeave, discusses the rapid evolution and challenges of AI infrastructure in response to the burgeoning demand for AI technologies. CoreWeave, a company initially rooted in the cryptocurrency mining sector, has pivoted towards providing specialized cloud infrastructure for AI and parallelizable workloads, distinguishing itself from traditional cloud services designed for serializable tasks. McBee highlights the company's significant growth, with plans to expand their data center operations across North America, underscoring the insatiable demand for AI compute resources. The conversation delves into the economics of GPUs, particularly NVIDIA's H100s, which are central to AI model training and inference. McBee explains the capital-intensive nature of setting up and operating data centers equipped with these GPUs, emphasizing the importance of efficient power use and innovative cooling techniques to manage the substantial heat generated by these systems. Despite the high costs associated with this infrastructure, the demand from companies seeking to train AI models and perform inference at scale continues to surge, with CoreWeave's services being fully booked. McBee also touches on the broader implications of AI infrastructure development, including the potential for AI to revolutionize various sectors such as advertising and healthcare. He predicts that generative AI will play a pivotal role in creating highly personalized and effective advertising, leveraging vast amounts of data to tailor content to individual users. Furthermore, McBee envisions AI contributing significantly to advancements in healthcare, although he anticipates this will become more pronounced in the latter half of the decade. Throughout the discussion, McBee underscores the challenges of scaling AI infrastructure to meet the explosive growth in demand. He points out the limitations of existing cloud infrastructure, which was not designed for the parallelizable workloads that AI requires, and the necessity of building new, specialized data centers. The conversation concludes with McBee expressing optimism about the future of AI and its capacity to drive innovation across various industries, despite the logistical and economic hurdles that lie ahead.

In episode 1924 of "This Week in Startups," the focus is on the state of modern answer engines, AI demonstrations, and the evolving landscape of AI technology. The episode begins with a discussion on Taylor Swift, humorously questioning her status as an icon or overrated artist, which transitions into a broader conversation about the capabilities of AI in creating content, including video, audio, and scripts. The hosts express amazement at the rapid advancements in AI, predicting that if AI can generate a nine-second clip today, it could potentially produce longer, more complex content in the future. The episode also highlights various AI-powered platforms and services, such as Vanta for SOC 2 compliance, Eight Sleep for improved sleep through technology, and the Equinix Startup Program for hybrid infrastructure solutions. The hosts delve into AI Tuesdays, emphasizing the importance of AI in startups and the tech industry at large. They discuss Grok Cloud's achievements in AI inference, offering insights into the cost-effectiveness and efficiency of AI models like Mixtral and GPT-4. A significant portion of the episode is dedicated to exploring modern answer engines like you.com and perplexity.ai, comparing their performance to Google's search engine. The hosts critique these platforms for their ability to provide timely and accurate information, especially in dynamic scenarios like sports odds or stock prices. They ponder the future of search, suggesting that voice and AI could revolutionize how users interact with information, potentially challenging Google's dominance in the search market. The episode also touches on ethical considerations and the potential for AI to influence human behavior, from customer support to more sinister applications like radicalization. The discussion on Hume.ai, an AI that understands human emotions, underscores the potential for AI to enhance human-AI interactions but also raises concerns about privacy and manipulation. Overall, episode 1924 of "This Week in Startups" offers a comprehensive look at the current state and future prospects of AI, from answer engines and cost implications to ethical considerations and the transformative potential of AI in various domains.

In this episode of the Liquidity Podcast, titled "Dave McClure & Jordan Stein on YC, Spray and Pray, and Games VCs Play," the discussion kicks off with a look at Reddit's IPO and its significance as another Y Combinator (YC) company going public. Dave McClure, founder of 500 Startups, shares his insights on YC's position in the industry and reflects on the "spray and pray" investment strategy, which he and YC were both criticized for. McClure explains how this approach allowed for a wide net of investments, leading to several big wins despite a high failure rate. He also touches on the importance of investing in diverse founders and the global reach of their investments. The conversation then shifts to the current investment climate, with Carta reporting a spike in capital call requests, indicating a return of VC bullishness. Jordan Stein from Crescent Partners discusses the slight increase in capital calls they've observed and attributes it to the growing interest in AI investments and a stabilizing market. Dave McClure talks about his fund, Practical VC, which focuses on secondary market investments, providing liquidity in a market where IPOs and acquisitions have slowed down. The podcast also delves into the challenges and strategies of running accelerators and seed funds. Jason Calacanis from The Launch Fund highlights the importance of taking massive risks on unique ideas to find breakout successes. He discusses the evolution of his investment strategy, focusing on identifying definitive winners and the decision-making process behind follow-on investments. Towards the end, the guests share their top three recent investments, highlighting companies like Mercury, Gropius (greener.com), and RecargaPay. They emphasize the potential in specific sectors like banking for startups, prefab multifamily housing automation, and payments platforms in Latin America. The episode wraps up with a discussion on the future of venture capital, the importance of portfolio strategy, and the value of learning from past investment decisions. The conversation underscores the dynamic nature of the VC industry and the continuous search for innovative ways to support startups and generate returns.

In episode 1922 of "This Week in Startups," host Jason Calacanis delves into the fascinating world of empathic AI with Alan Cowen, CEO and Chief Scientist at Hume AI. The episode explores Hume AI's mission to bridge the gap between artificial intelligence and emotional intelligence, aiming to understand and optimize AI for human well-being. Cowen explains that much of human expression, such as laughter, sadness, or pain, is conveyed through voice and facial expressions rather than language alone. Hume AI has developed models that better understand these expressions and integrate them into large language models, allowing AI to adapt and respond to human emotions more effectively. The conversation also touches on the potential applications of empathic AI, from improving customer service to enhancing therapy and coaching. Cowen shares insights into the technology behind Hume AI, including its ability to analyze voice and facial expressions in real-time to gauge emotions. This capability opens up new possibilities for AI to interact with humans in a more nuanced and empathetic manner. However, the episode also addresses the ethical considerations surrounding empathic AI. Cowen emphasizes the importance of using this technology to enhance human well-being rather than manipulate emotions for commercial or political purposes. He discusses Hume AI's ethical guidelines and the role of the Human Initiative in ensuring that AI applications prioritize users' interests and satisfaction. Throughout the episode, Calacanis and Cowen engage in demonstrations of Hume AI's technology, showcasing its ability to understand and respond to emotional cues. These demos highlight the potential of empathic AI to transform human-AI interactions, making them more personal and meaningful. In summary, episode 1922 of "This Week in Startups" offers a deep dive into the world of empathic AI with Hume AI's Alan Cowen. The discussion covers the technology's capabilities, applications, and ethical considerations, providing a comprehensive overview of how AI can be optimized for human well-being.

In episode 1921 of "This Week in Startups," titled "Ask Jason LIVE! Unpacking Startup Strategies with Real-Time Q&A," Jason Calacanis hosts a live Q&A session, addressing a wide range of questions from startup founders and entrepreneurs. The episode covers various topics, including investment strategies, the future of seed stage investment structures, and the impact of AI on startups. One of the key discussions revolves around how seed stage investment structures, like convertible notes and safes, might adapt to startups aiming for sustainable growth with minimal external funding. Jason highlights the importance of these financial instruments in early-stage investing and shares insights on how founders can navigate the complexities of fundraising and investor relations. The conversation also delves into the role of AI in startups, with Jason emphasizing that AI is becoming a fundamental technology that will be incorporated into virtually every product. He advises founders to be clear about how AI is used in their products and to consider its potential in improving operations, product features, and future opportunities. Jason encourages startups to think about AI not just as a buzzword but as a tool that can provide significant advantages, such as data analysis and customer support enhancements. Throughout the episode, Jason offers practical advice on building a startup, from understanding customer needs and defining a clear value proposition to efficiently using technology like AI. He stresses the importance of execution, building a delightful product, and being open to learning and adapting based on market feedback. Listeners also get a glimpse into the experiences of founders who have participated in Founder University, a program run by Jason's team. The program is praised for helping participants refine their business ideas, accelerate their startup's growth, and prepare for fundraising. Overall, the episode provides valuable insights and guidance for entrepreneurs at various stages of their startup journey, highlighting the importance of adaptability, customer focus, and leveraging technology like AI to build successful businesses.

In episode 1920 of This Week in Startups, titled "AI Demos Gatekeep, Deepgram, Claude 3 Haiku and more!", the hosts explore various AI tools and their applications in analyzing conversations, generating content, and interpreting images. The episode kicks off with a discussion on Deepgram's new tool that can perform intelligent analysis of audio files, breaking down the intents of a conversation and providing sentiment analysis. This tool is praised for its potential in understanding how conversations are perceived, which could be particularly useful in evaluating sales calls or investor pitches. The conversation then shifts to Grok, a platform that has seen rapid growth in developer engagement, with over 60,000 developers and 15,000 apps utilizing its capabilities. The hosts discuss the importance of space for startups, reflecting on how the availability of affordable spaces can inspire innovation and community building. They also touch upon the significance of AI in enterprise, highlighting a report by A16Z that shows a significant increase in enterprise spending on AI technologies. The episode also delves into the challenges and opportunities presented by generative AI, with a focus on the need for accuracy in AI-generated content. The hosts express concerns about the potential for AI to disseminate incorrect information, emphasizing the importance of developing AI models that prioritize accuracy. They explore various AI demos, including Gatekeep.ai, which generates explainer videos and summaries from text prompts, and Deepgram's tool for analyzing and summarizing audio content. Despite recognizing the innovative potential of these tools, the hosts remain cautious about their current limitations, particularly in terms of accuracy. Anthropic's Cloud 3 Haiku model is also showcased, demonstrating its ability to quickly and accurately describe images in detail, including estimating the contents of a food bowl. This example illustrates the advancements in multimodal AI and its potential to simplify complex tasks, such as calorie counting or content creation, into more manageable processes. Throughout the episode, the hosts emphasize the transformative impact of AI on various industries, from startups to enterprise-level companies. They encourage entrepreneurs and developers to explore and build with AI technologies, while also advocating for a focus on accuracy to ensure the reliability of AI-generated content. The episode concludes with a call to action for the AI community to continue pushing the boundaries of what's possible with AI, while also addressing the challenges that come with it.

In episode 1919 of the Liquidity Podcast, titled "Kleiner Perkins and Penn’s Endowment on India, loss ratios, and the rise of secondaries," the discussion kicked off with a focus on the increasing interest of Middle Eastern sovereign wealth funds in India, highlighting a significant investment in Pocket FM by Abu Dhabi's sovereign wealth fund, Adia. Mamoun Hamid from Kleiner Perkins expressed bullish sentiments on India, citing its large population, burgeoning middle class, and infrastructure improvements as key growth drivers. The conversation then shifted to the challenges and opportunities within the Indian ecosystem, including the importance of strong leadership and the potential for Indian startups to address local markets effectively. The podcast also delved into Carta's data on the 2018 startup class, revealing a high mortality rate among startups, with over 49% already dead and only 0.2% going public. This led to a discussion on the importance of loss ratios in venture capital and the strategies employed by venture firms to manage their portfolios effectively. The rise of secondaries as a significant asset class was another focal point, with Thomas Scriven from the University of Pennsylvania Endowment providing insights into how sovereign wealth funds and endowments view this trend. The episode concluded with each guest sharing their recent investments, highlighting a strong focus on AI and healthcare. Mamoun Hamid discussed Kleiner Perkins' investments in companies like Glean, Ambience Healthcare, Harvey, and Codium, emphasizing the potential of AI to enhance productivity in various sectors. Jason Calacanis shared his investments in early-stage startups through his accelerator and Foundry University programs, focusing on areas such as affiliate marketing for SaaS, AR for job sites, and podcasting intelligence. Thomas Scriven, while maintaining confidentiality on specific investments, mentioned Penn’s Endowment's interest in early-stage, capital-efficient models, particularly in life sciences and digital health, driven by strong strategic networks. Overall, the episode provided a comprehensive overview of the current trends in venture capital, with a particular emphasis on the opportunities in India, the challenges of achieving successful exits in a tight IPO and M&A market, and the strategic approaches to portfolio management and investment in emerging technologies.

In episode E1918 of "This Week in Startups," the hosts delve into Microsoft's strategic move of acquiring the team from Inflection, a company with deep ties to AI innovation and founded by a co-founder of DeepMind. This acquisition is seen as a bold and cutthroat strategy by Microsoft's CEO, Satya Nadella, to consolidate talent and eliminate competition in the AI space. Inflection, known for its significant investment in NVIDIA hardware and its ambition to compete with OpenAI, was in the process of developing a chatbot and a foundational model before Microsoft intervened. The episode also touches on the broader implications of such acquisitions in the tech industry, including the potential for regulatory scrutiny and the strategies companies might employ to navigate these challenges. The podcast also highlights various AI demos, showcasing the advancements and capabilities of different AI models. Suno, an AI model capable of generating music based on user descriptions, impresses with its ability to create songs that closely match specified genres and styles, even without directly referencing specific artists. Another AI model, Command-R from Cohere, demonstrates its ability to augment search results with sourced information, providing a more transparent and verifiable output compared to other models. DeepSeek, a Chinese AI model, showcases its ability to interpret and describe images with remarkable accuracy, including understanding context and actions depicted in photographs. Throughout the episode, the hosts emphasize the rapid progress in AI technology, highlighting the potential for these models to transform various industries and creative processes. They also discuss the ethical and legal considerations surrounding AI, particularly in terms of copyright and data usage, suggesting a need for clear guidelines and fair practices as AI continues to evolve. The episode concludes with a reflection on the future of AI and its implications for both businesses and consumers, underscoring the importance of staying informed and engaged with the latest developments in this fast-moving field.

In this episode of "This Week in Startups," Jamie Rhode of ScreenDoor discusses the importance of optimizing fund structure and finding a good GP (General Partner) market fit. ScreenDoor is a fund of funds that aims to support underrepresented voices in the venture capital space by backing investors on their first rounds of investment. Jamie emphasizes the need for venture managers to explore untapped potential in the ecosystem by investing in overlooked areas and founders. This approach can break the virtuous cycle and channel more capital towards innovative ideas and sectors. Jamie highlights the significance of early-stage venture being parallel driven, where the edges or tails of investment opportunities drive returns. To capture these opportunities, portfolio construction for LPs (Limited Partners) is crucial. It involves investing in enough fund managers to cover a broad swath of diversification, ensuring exposure to potential big winners while acknowledging the high risk involved. Jamie suggests that a portfolio should include a mix of investments in new sectors and emerging managers who bring fresh perspectives and networks to the table. The discussion also covers the importance of GP market fit, where a GP's expertise and strategy align with their long-term vision and competitive advantage. Jamie points out that successful venture fund managers need to stay grounded in seed-stage investing, keeping fund sizes small to focus on high-potential startups. This approach requires discipline and a long-term perspective, as significant returns may take years to materialize. Jamie and the host also delve into the challenges of fund construction, debating the merits of different strategies such as concentration in winners versus a broader spread of investments. They discuss the importance of ownership percentage in startups and the decision-making process for follow-on investments. Jamie advocates for a data-driven approach to these decisions, emphasizing the need for venture fund managers to be disciplined and strategic in their investment choices. Overall, the episode sheds light on the complexities of venture capital investing, the importance of supporting underrepresented voices, and the strategies that can lead to successful outcomes for both investors and startups. Jamie's insights offer valuable guidance for both emerging and established venture fund managers navigating the dynamic venture capital landscape.

In this episode of "This Week in Startups," host Jason Calacanis interviews Arvind Jain, the co-founder of Glean, a company aiming to revolutionize how enterprises search and utilize their internal data through AI-powered tools. Glean is described as an internal version of Google or ChatGPT for companies, enabling employees to ask any question and receive answers based on the company's collective knowledge and data. This AI-powered search engine and assistant are designed to enhance work productivity by making it easier for employees to find the information they need across various departments and data sources. Glean primarily targets CIOs for company-wide deployment, with engineers, support staff, and salespeople being the top users. The product integrates with existing systems like Intercom, Zendesk, and Slack, acting as a connective tissue that links knowledge across different platforms. It helps employees find answers not just in knowledge articles but also in Slack conversations, JIRA issues, and even identifies relevant experts within the company. Glean's technology is permissions-aware, ensuring that employees only access information they are authorized to see, addressing potential confidentiality concerns. The discussion also touches on the future of AI in the enterprise, with Jain predicting that the majority of AI applications will be based on open-source models due to the inherent advantages of open-source development. He emphasizes the importance of data privacy and governance, especially in the context of using AI within enterprises. Jain shares insights into the challenges and opportunities presented by AI, including the potential for AI to significantly impact employee productivity and organizational knowledge management. Jain's vision for Glean is to make it an indispensable assistant for every employee, simplifying the process of finding and utilizing information across an organization. He believes that AI has the potential to transform work productivity but also acknowledges the complexities and challenges involved in achieving this goal. The conversation concludes with a discussion on the pricing model for Glean, which is based on a per-seat basis, making it accessible to both mid-sized companies and large enterprises.

In episode 1915 of This Week in Startups, the hosts delve into a variety of AI-related topics, starting with a significant bet on the availability of humanoid robots by January 1, 2027. They speculate that by this date, humanoid robots will be purchasable and deliverable to homes, with a bet involving the purchase of such a robot for the winner, setting the stakes high at $20,000. This discussion leads to broader conversations about the advancements in AI and robotics, including the open-sourcing of Grok by Elon Musk and its implications for the AI community. The episode also touches on the legal and ethical considerations of using open web data for training AI models, highlighting the complexities of intellectual property rights in the AI era. The podcast further explores the integration of AI in various applications, particularly focusing on Figure AI's development of a humanoid robot capable of interacting with humans through voice recognition and reasoning. This segment emphasizes the potential of combining AI language models with robotics to perform tasks and make decisions based on human-like reasoning. The discussion extends to the broader implications of AI in transforming business operations, suggesting a future where AI agents, referred to as "maestros," could manage multiple aspects of a business, leading to highly efficient and lean operations. Additionally, the episode covers the collaboration between Apple and Google regarding the licensing of Gemini, Google's generative AI models, for use in iPhone software. This potential partnership is analyzed in the context of the existing search deal between the two tech giants and the strategic importance of AI in maintaining competitive advantages. The hosts speculate on how this collaboration could shape the future of AI integration in consumer technology, emphasizing the significance of AI in enhancing user experiences and the potential regulatory scrutiny such partnerships might attract. Overall, episode 1915 of This Week in Startups presents a comprehensive discussion on the current state and future prospects of AI and robotics, highlighting significant bets, technological advancements, and strategic partnerships that could define the trajectory of AI integration in various sectors.

In the podcast episode titled "2024 VC trends, portfolio construction, & more with Churchill's Raja Doddala," the discussion delves into the current state and future trends of venture capital, focusing on the implications of recent market adjustments for investors and startups. The conversation begins with an observation of the significant decrease in cash being invested in pre-seed and seed stages, marking a 13-quarter low. This shift is attributed to a collective realization within the industry of the need to return to more sustainable investment practices after a decade of rapid and often reckless capital deployment. The episode highlights the normalization of deal counts and values, suggesting a return to pre-pandemic levels and a departure from the frenzied investment pace of 2021 and 2022. This adjustment is seen as a healthy correction, allowing for more disciplined investment strategies and a focus on achieving meaningful milestones before seeking further funding. The discussion also touches on the specific challenges and opportunities presented by the burgeoning field of artificial intelligence (AI), noting that while AI has the potential to drive significant innovation, there is also a risk of disillusionment due to inflated expectations and premature valuations. Raja Doddala, representing Churchill Asset Management, shares insights into portfolio construction and the importance of diversification across different stages of venture capital. He emphasizes the value of a disciplined approach to investing, particularly in the context of the current market correction. The conversation also explores the dynamics of secondary sales and the strategic management of equity positions in startups, underscoring the importance of liquidity for both investors and founders. Overall, the episode provides a comprehensive overview of the current venture capital landscape, offering valuable perspectives on how investors and startups can navigate the evolving market conditions. The emphasis on disciplined investment, the potential impact of AI, and the strategic considerations surrounding portfolio construction and liquidity management are highlighted as key factors shaping the future of venture capital.

In episode 1913 of "This Week in Startups," host Jason Calacanis interviews Justin Kwong, the founder of Malengo, a company that simplifies the process of developing and scaling fashion products. Malengo went through Calacanis's accelerator program and has seen significant success since. The company's mission is to make it easy for brands or creators to develop, finance, and scale an apparel line, handling the complex parts of the manufacturing process and using technology to automate parts of it, thus reducing the typical development time from months to weeks. Malengo's service is particularly beneficial for influencers or brands looking to create high-quality merchandise, offering a tech-enabled dashboard for product development. This includes everything from initial design to sample creation, with a focus on transparency and efficiency throughout the process. The company charges for development packages to create samples, with prices ranging from $300 to $500, which is significantly lower than traditional methods. Malengo then makes a margin on bulk production orders, providing a cost-effective solution for brands to develop high-quality apparel. The episode also delves into the broader implications of automation and efficiency in startups, with Calacanis discussing his new ADD (Automate, Deprecate, Delegate) framework for improving company efficiency. He emphasizes the importance of startups being resourceful and leveraging technology to streamline operations, which is particularly relevant in the context of Malengo's business model. Overall, the episode highlights Malengo's innovative approach to fashion product development and the broader trends in startup efficiency and automation. Justin Kwong's insights into the challenges and solutions in the apparel industry, combined with Calacanis's discussion on productivity, offer valuable perspectives for entrepreneurs and startups looking to optimize their operations in any sector.

In episode 1912 of This Week in Startups, titled "AI Demos Claude 3's Opus, Mistral, Groq Playground, EMO by Ali Baba," hosts and guests delve into the latest advancements and demonstrations in artificial intelligence. The episode kicks off with a discussion about the acquisition of Definitive Intelligence by Grok (G-R-O-K-U), highlighting the significant impact and the future plans for the company under Grok's umbrella. The conversation transitions into the capabilities of Grok's AI chip and its cloud service, emphasizing the high-speed inference engine that allows developers to build new experiences with super low latency and high throughput. The episode further explores various AI models, starting with Claude 3's Opus, which is praised for its reasoning capabilities and the provision of a prompt library to optimize model interaction. The discussion also covers Mistral's AI model, Mistral Large, which excels in writing code and solving mathematical problems, showcasing its functionality and contribution to the open-source ecosystem. Another highlight is Grok's playground, a platform that enables developers to use different models and integrate AI into their applications, demonstrating its utility in educational settings by creating study plans and lessons tailored to individual interests. The episode concludes with a demonstration of an AI model developed by Alibaba's Institute for Intelligent Computing, which animates static photos to sync with songs, showcasing the potential for creating lifelike animations from still images. Throughout the episode, the hosts and guests provide insights into the implications of these AI advancements for developers, the open-source community, and society at large, emphasizing the importance of competition and innovation in the AI space.

In episode 1911 of "This Week in Startups," host Jason Calacanis interviews Josh Mohrer, the former General Manager of Uber in New York City and now an angel investor and founder of a new AI startup. The episode delves into Mohrer's journey from leading Uber's expansion in New York, one of its first and most vibrant markets, to embracing the role of a solo founder in the AI era. Mohrer shares insights into Uber's early days, including the company's management style under Travis Kalanick, the challenges and strategies in dealing with regulatory hurdles, and the evolution of Uber's business model towards profitability. Mohrer's new venture, an AI note-taking app called Wave AI, is inspired by the potential of AI to empower individuals to build and scale businesses single-handedly. He discusses the development process, leveraging AI tools like ChatGPT to code and build the app himself, despite not being a developer by trade. The app, which transcribes and summarizes audio recordings, has seen rapid growth and positive user feedback, highlighting the demand for efficient, AI-powered productivity tools. The conversation also touches on broader themes, such as the impact of AI on employment, the future of work, and the potential for individuals to create niche, yet highly successful businesses with the help of AI. Mohrer's journey exemplifies the shift towards a new era of entrepreneurship, where solo founders can leverage AI to bring their visions to life without the need for large teams or significant capital. Throughout the episode, Calacanis and Mohrer explore the implications of AI's advancement for startups, the workforce, and society at large. They discuss the potential for AI to democratize the startup ecosystem, enabling more people to participate in innovation and entrepreneurship. The episode concludes with a reflection on the transformative power of AI and the exciting possibilities it holds for the future of business and creativity.

In this episode of "This Week in Startups," host Jason Calacanis interviews Jonathan Neman, co-founder and CEO of Sweetgreen, a fast-casual restaurant chain specializing in salads and now offering a broader menu including protein plates and warm bowls. Sweetgreen, which started in 2007 in Georgetown, Washington D.C., has expanded to over 225 locations across the United States. The company has been at the forefront of innovation in the restaurant industry, leveraging technology such as robotics, pay loyalty, and data-driven menu optimizations to enhance customer and team member experiences. Approximately 60% of Sweetgreen's business comes through its mobile app, highlighting the company's digital-first approach. Neman discusses the challenges of maintaining a high-quality customer experience amidst labor shortages and the increasing minimum wage. Sweetgreen's solution involves using technology to improve both the customer and team member experience, focusing on cooking and hospitality while automating other aspects of the service. The company has introduced "infinite kitchen technology" in some stores, where robotics handle most of the food assembly, significantly increasing efficiency and accuracy while allowing human employees to focus on culinary aspects and customer service. The conversation also touches on Sweetgreen's commitment to health and sustainability, addressing the obesity and metabolic dysfunction crisis in the United States by offering nutritious, fresh food. Neman criticizes the current food system, heavily subsidized towards corn and soy, which contributes to the prevalence of processed foods. Sweetgreen aims to be part of the solution by working directly with farmers to promote organic agriculture and by making healthy food sexy and cool through marketing and cultural relevancy. Neman shares his vision for Sweetgreen to offer personalized health solutions through its menu, leveraging AI to recommend dishes based on individual customer preferences and nutritional goals. He also discusses the challenges and opportunities of expanding Sweetgreen's reach to underserved communities, emphasizing the importance of education, access, and affordability in promoting healthier eating habits. Lastly, the episode covers Sweetgreen's approach to delivery services and the potential of automation and robotics to lower costs and make healthy food more accessible. Neman expresses skepticism towards mock meats, advocating for real, unprocessed foods as the key to a healthy diet.

In the episode titled "Business Breakdowns: Moneyball's First-Principles, Paradigm Shifts, and ‘Burn the Boats’" from "This Week in Startups," the hosts delve into the story of "Moneyball," a film that encapsulates the application of first-principle thinking and statistical analysis in baseball, a domain traditionally dominated by intuition and experience. The narrative revolves around Billy Beane, the general manager of the Oakland A's, who, constrained by a limited budget, adopts a revolutionary approach to assembling a competitive team. By focusing on undervalued players with high on-base percentages, Beane challenges the conventional wisdom of baseball scouts and executives, sparking a paradigm shift in the sport. The episode highlights the resistance Beane faces from the old guard, including scouts and his team manager, Art Howe, who are deeply entrenched in traditional baseball ideologies. This resistance is emblematic of the broader challenges innovators face when introducing new methodologies in any field. The hosts discuss the importance of adaptability, persistence, and commitment to a vision, even in the face of skepticism and opposition. Beane's strategy, although initially met with criticism and doubt, ultimately leads to an unprecedented winning streak for the A's, proving the efficacy of his approach. The discussion also touches on the broader implications of "Moneyball's" philosophy beyond baseball, drawing parallels to startup culture and business strategy. The hosts emphasize the value of data-driven decision-making, the willingness to challenge established norms, and the significance of identifying and leveraging undervalued assets, whether in sports or in the business world. The episode concludes with reflections on the personal and professional growth of Beane, who, despite receiving lucrative offers from wealthier teams, chooses to stay with the A's, underscoring the intrinsic rewards of pioneering change and redefining success on one's own terms. Throughout the episode, the hosts engage in a lively and insightful discussion, interspersed with clips from the film and interviews, to illustrate the transformative power of embracing innovation and the courage required to "burn the boats" and fully commit to a new direction. "Moneyball" serves as a compelling case study in overcoming adversity, challenging the status quo, and the relentless pursuit of excellence through unconventional means.

In the episode titled "Mission Over Money & The State of Defense Tech with Palantir's Shyam Sankar," the discussion delves into the evolving landscape of defense technology, emphasizing the critical role of adaptability and speed in modern warfare. Shyam Sankar, the CTO of Palantir and the company's 13th employee, shares insights into the company's journey and its pivotal role in defense tech, highlighting the importance of the OODA loop (observe, orient, decide, act) in gaining a decision advantage in conflicts. Initially developed for dogfighting in airplanes, the concept underscores the necessity of iterating on defense technologies at the pace of the fight to outmaneuver adversaries. The conversation also touches on the challenges and opportunities in the defense tech sector, including the need for re-industrialization in the United States to maintain a competitive edge. Sankar discusses the historical context of America's industrial mobilization during World War II and draws parallels to the current need for innovation in defense manufacturing. He advocates for leveraging America's software prowess to develop more efficient and adaptable defense solutions, emphasizing the potential of integrating artificial intelligence and autonomous systems into military operations. Furthermore, Sankar addresses the geopolitical implications of emerging technologies, such as hypersonic glide vehicles and the contested domain of space. He stresses the importance of taking adversaries like Russia and China seriously, noting their advancements in anti-satellite capabilities and the strategic significance of space in future conflicts. The discussion also explores the impact of drones and other unmanned systems on the battlefield, highlighting the shift towards cheaper hardware with sophisticated software coordination. In conclusion, the episode underscores the critical role of defense technology in national security and the importance of innovation, adaptability, and strategic investment in maintaining America's military edge. Sankar's insights offer a glimpse into the challenges and opportunities facing the defense tech sector, emphasizing the need for mission-driven efforts over purely financial motivations.

In the episode titled "Sequoia & Union Square Returns, the Post-AI Labor Market, and the Return of SF E1907" of the Liquidity Podcast, the discussion kicks off with the revelation of venture capital returns, including those from Sequoia Capital and Union Square Ventures, as disclosed by the University of Texas Endowment following a public records request. Union Square Ventures stands out with a 9.14x cash-on-cash return, indicating a strong performance across its funds. The conversation then shifts to the broader VC market, highlighting the significant variance in returns between the top and bottom quartiles of firms and the impact of market timing on investment success. The panelists, including Eric Thornburg, Guy Perlmutter, and Jason Calacanis, delve into the strategies employed by venture capital firms when exiting investments, emphasizing the importance of a clear and consistent policy. They discuss the potential for AI to reshape the labor market, with a focus on the recent trend of companies operating with significantly reduced staff, inspired by Elon Musk's drastic workforce reduction at Twitter. This shift towards automation and efficiency is seen as a potential turning point, signaling a move away from the Silicon Valley culture of large teams and lavish employee perks. The conversation also touches on the geographic dynamics of the startup ecosystem, with a particular focus on the return of founders and investors to San Francisco after a pandemic-induced exodus to cities like Miami. Despite the challenges posed by local governance, San Francisco's strong network effects and the renewed political engagement of the tech community are highlighted as key factors driving this return. Lastly, the episode features a segment where the panelists share their recent investments, showcasing a diverse range of startups from various sectors, including AI-driven platforms for engineering communication, nuclear energy, cybersecurity, manufacturing efficiency, consumer electronics, sports training, and code optimization. These investments reflect the panelists' optimism about the potential for innovation across different industries and the ongoing evolution of the venture capital landscape.

In the episode "Unlocking the Power of Data with VenCap’s David Clark" of "This Week in Startups," David Clark, the Chief Investment Officer at VenCap, shares his insights from three decades of investing in venture funds. VenCap, a UK-based fund of funds and investment advisor, has made around 500 fund investments since its inception in 1987, focusing on a select group of 12 to 15 managers in recent years. Clark discusses the importance of recognizing and investing in top-performing companies, noting that the best managers understand when to let their investments run to achieve fund-returning outcomes. Clark elaborates on the venture capital industry's power law dynamics, where a small number of exits generate the majority of value. He emphasizes the challenge of consistently accessing these top-performing companies and highlights the concentrated group of managers who have been successful in doing so. The discussion also covers the importance of fund size and the relationship between exit size, fund ownership, and the ability to generate significant returns. The conversation shifts to the challenges facing the venture capital industry, including the influx of new investors and the difficulties of managing portfolio companies through tough times. Clark stresses the importance of portfolio management, where the best-performing funds allocate more capital to their best-performing companies and less to their worst-performing ones. He also touches on the impact of fund size on performance, advocating for right-sizing fund sizes to optimize for fund-returning outcomes. Throughout the episode, Clark and the host discuss various aspects of venture capital investing, including the importance of succession planning in venture firms, the dynamics of premium carry and fees, and the resurgence of the J-curve in venture investing. The episode concludes with a reflection on the importance of founders and investors recognizing when to persevere with a company and when to make the difficult decision to shut it down, underscoring the complex and challenging nature of venture capital investing.

In episode 1905 of This Week in Startups, titled "Ask Jason LIVE! Returns to TWiST!", Jason Calacanis revives the live call-in format, allowing entrepreneurs and startup enthusiasts to ask him questions directly. The episode kicks off with a discussion about the value of pursuing a master's in fintech, with Jason advising the caller, Tommy, against spending $40,000 on the degree. Instead, he suggests gaining practical experience by working at a startup, emphasizing the importance of being in a tech hub like Silicon Valley for networking and opportunities. Jason then addresses common mistakes made by first-time founders, highlighting the pitfalls of developing a product without customer input and the necessity of having a team with the right skills. He advises founders to focus on solving real customer problems and to ensure their teams are capable of building the product. The conversation shifts to a startup that offers AI-powered sourcing and interviewing for recruiters. Jason explores the startup's approach to solving a genuine market need and the importance of validating the product with customers willing to pay for it. He stresses the significance of finding a product-market fit and the challenges of ensuring the solution addresses a critical need for its users. Jason also discusses the challenges faced by startups during tough economic times, using the example of Cafe X, a robotic coffee machine startup. He emphasizes the resilience required to survive and eventually thrive, suggesting that startups focus on solving acute problems for a small number of customers to attract angel investors and seed investors. The episode concludes with advice on raising funds before achieving product-market fit. Jason underscores the importance of having a strong team, a clear vision, and early customer traction. He suggests that joining an accelerator or incubator can provide validation and attract seed funding, even for startups outside major tech hubs. Throughout the episode, Jason provides practical advice on navigating the startup landscape, emphasizing the importance of customer feedback, team composition, and strategic focus for early-stage companies.

In episode E1904 of "This Week in Startups," hosts Jason Calacanis and Sunny Madra delve into the latest advancements and discussions surrounding artificial intelligence (AI), with a particular focus on Google's AI emergency, Apple's AI developments, and the impressive demonstrations of Sora by OpenAI. The episode kicks off with high praise for Sora, labeling it as one of the best AI demos in history and awarding it an A+ rating. The hosts are astounded by the quality and capabilities demonstrated by the OpenAI team, particularly in creating long-duration content with complex camera movements and realistic details. The conversation then shifts to a technical explanation of why AI models, like Google's large language models, can produce distorted or biased responses. Madra breaks down the three main factors that influence an AI model's responses: the training data, reinforcement learning by humans, and guardrails. He explains how these elements can skew a model's output, using Google's "super woke" responses as an example of potential overreach in implementing guardrails. The discussion highlights the importance of transparency and the potential for open-sourcing guardrail models, as seen with Meta's LamaGuard, to foster better understanding and regulation of AI behavior. The episode also touches on Apple's subtle yet significant AI moves, as demonstrated by new features in iPhoto that can identify and categorize images with remarkable accuracy. Calacanis and Madra are impressed by Apple's ability to integrate AI seamlessly into its products, providing users with powerful tools without much fanfare. Throughout the episode, various AI tools and demos are reviewed, including explorer.globe.engineer for research assistance and reika.ai for multimodal vision models. Each tool is evaluated for its potential impact and utility, with the hosts expressing excitement about the future possibilities these AI advancements hold. In closing, the episode reflects on the broader implications of AI on content creation, from movies to music, and the potential for AI to streamline processes, enhance creativity, and even challenge traditional production methods, as evidenced by Tyler Perry's decision to pause a significant studio expansion in light of AI developments like Sora. The hosts conclude with anticipation for the continued evolution of AI and its growing influence across various industries.

In this episode of the Liquidity Podcast, the discussion kicks off with Reddit's announcement of going public through its S1 filings, revealing over $804 million in revenue. Brian Rosenblatt, a former Reddit employee, shares his bullish outlook on the company, attributing its success to a combination of product improvements, advertiser education, and community engagement. The conversation then shifts to the broader implications of AI in the business world, with a focus on the distinction between consumer and enterprise AI applications. The panelists explore the challenges and opportunities presented by AI, emphasizing the importance of unique data and the potential for AI to revolutionize various industries. The episode also introduces a new corporate venture capital fund led by Sam Altman, highlighting its strategic investments in AI startups and the potential conflicts and advantages of corporate VC funding. The discussion delves into the dynamics of portfolio construction, the role of emerging managers in identifying unicorn companies, and the strategic importance of follow-on investments in successful startups. The panelists share their latest investments, showcasing a diverse range of startups leveraging AI and other technologies to innovate in their respective fields. From healthcare advocacy platforms and advertising solutions for YouTubers to AI-driven developer vetting services and a comprehensive platform for designing and manufacturing custom products, these investments reflect the panel's optimism about the transformative potential of AI and other emerging technologies. Throughout the episode, the conversation underscores the complexities and rapid evolution of the AI landscape, the strategic considerations for investors and startups navigating this space, and the enduring importance of building relationships and understanding market dynamics in the venture capital ecosystem.

The podcast episode titled "Startup Pitches! LAUNCH Accelerator Demo Day E1902" showcases seven startups presenting their innovative solutions to a panel of guest investor judges. The startups, part of the Launch Accelerator Cohort 30, are introduced as having product velocity, great product design, and a focus on customer obsession. The founders have three minutes to pitch their ideas, followed by a Q&A session with the investors. The episode highlights the importance of having multiple founders with diverse skills, such as a developer, UX designer, and a growth hacker, to drive product development and customer engagement. The startups featured in the episode span various industries, including fashion manufacturing, podcast automation, affiliate marketing, event and social platform management, AI for tax professionals, creative funding platforms, and adaptive learning environments for exams. Each startup presents a unique solution to a specific problem, demonstrating the potential for growth and market impact. For instance, Malengo offers a managed marketplace for fashion product development, Podcast AI automates podcast production, and TaxGPT provides an AI co-pilot for tax professionals. Investor judges, representing venture capital firms such as Tenacity Venture Capital, Outlander, 10.1.10, Costanoa, and Combative Ventures, provide feedback and select their top three startups based on the pitches. The episode concludes with a discussion on the challenges and opportunities for startups in raising funds, especially in a tough market. The importance of focusing on customer satisfaction and product development is emphasized as key to startup success. Overall, the episode showcases the diversity of innovation within the startup ecosystem and the critical role of accelerator programs in nurturing and presenting new ventures to potential investors. The interaction between the startups and the investor judges offers valuable insights into the investment decision-making process and the factors that contribute to a startup's attractiveness to venture capitalists.

In this episode of "This Week in Startups," the final season of the Angel series, host Jason Calacanis interviews Jordan Stein from Cresset Partners, focusing on the dynamics of venture capital (VC), the process of evaluating emerging managers, and the broader landscape of private investing. Cresset Partners is a multi-family office managing investments for ultra-high net worth individuals, dealing with various asset classes including real estate, private credit, private equity, and venture capital, which is highlighted as the most attractive yet challenging to access. Jordan Stein, the director of Cresset's Venture Capital Investment Practice, shares insights into why venture capital is a significant area for investment, emphasizing its historical returns and the difficulty of accessing top venture managers. He discusses the importance of investing with the best managers to ensure worthwhile returns, noting the barriers many family offices and high net worth individuals face in gaining access to top quartile or decile managers. Stein explains Cresset's approach to building a diversified asset allocation strategy, incorporating venture capital to meet the growth and preservation mandates of their clients. The conversation also delves into the peculiarities of the venture capital industry, including the challenges of engaging with venture capital firms and securing allocations in top-tier funds. Stein highlights the importance of relationships and networking within the industry to gain access to these funds, sharing Cresset's journey and strategy in establishing a credible presence in the venture capital space. He mentions Cresset's success in investing with renowned firms like Andreessen Horowitz, Founders Fund, and Cowboy Ventures, attributing this to the firm's growth and strategic acquisitions that enhanced their access and appeal to general partners (GPs). Furthermore, Stein and Calacanis discuss the impact of market conditions on venture capital fundraising and investment strategies, touching on the resilience of top venture firms despite economic downturns. They explore the concept of venture tourism, the challenges faced by new and less established venture firms, and the strategic importance of co-investment opportunities to blend down fees and generate faster distributions to paid-in (DPI). The episode also covers the democratization of venture capital, with both Stein and Calacanis expressing support for broader access to venture investments for accredited and potentially non-accredited investors. They critique the current regulatory environment that restricts access to venture capital investments, advocating for an accreditation test that could educate potential investors on the risks and rewards of venture capital. Overall, the podcast provides a comprehensive overview of venture capital dynamics, the evaluation of emerging managers, and the strategies employed by Cresset Partners to navigate the complex landscape of private investing.

In episode 1900 of This Week in Startups, host Jason Calacanis discusses the future of public safety through rapid drone response with Rahul Sidhu, CEO and co-founder of Aerodome. The conversation begins with a look at the venture capital interest in defense and public safety startups, highlighting a shift in Silicon Valley's investment landscape towards these sectors. Aerodome's innovative approach to public safety involves deploying drones to respond to 911 calls, providing rapid air support to police and fire departments. These drones can reach any location in a city within an average of 98 seconds, offering a scalable, faster, and more cost-effective alternative to traditional helicopter-based air support. Rahul Sidhu shares his background in both technology and public safety, including his experience as a police officer, firefighter, paramedic, and air support manager. He explains how the idea for Aerodome emerged during the COVID-19 pandemic when traditional air support units were grounded, and there was a need to do more with less. The company's drones are launched from central stations, where their batteries are swapped by robot arms, and they can be controlled remotely by operators at police or fire stations. The conversation also touches on public concerns regarding privacy, safety, and potential abuse of the technology. Sidhu emphasizes the importance of transparency and accountability, noting that the drones' video data can be subject to the Freedom of Information Act. He also discusses the legal framework within which the drones operate, comparing their use to that of helicopters in terms of case law and operational principles. Aerodome's business model is explored, with Sidhu explaining how the company offers a complete solution that includes software, hardware, and services for a subscription fee. This approach aims to democratize access to air support for public safety agencies that cannot afford traditional helicopter programs and to provide a more efficient and safer alternative for those that can. The episode concludes with a discussion on the potential for further automation and autonomy in drone technology, as well as the challenges and opportunities in selling to municipalities. Sidhu shares his personal journey and the impact of the All-In podcast on his thinking during a tumultuous period in 2020, highlighting the importance of intellectual honesty and problem-solving in both public safety and entrepreneurship.

In the episode titled "Mark Suster and Samir Kaji on the 2024 Venture Market, IPOs, and Secondaries" of the Liquidity Podcast, hosted by David Weisberg with guests Mark Suster, Samir Kaji, and Jason Calacanis, a comprehensive discussion unfolds around the venture capital landscape, focusing on the anticipated market trends for 2024, the state of IPOs, and the dynamics of secondary markets. The conversation begins with insights into IVP's new fund, highlighting the venture capital firm's consistent fundraising strategy and the current state of Series B and C funding rounds. The discussion reveals a market adjustment with reduced valuations and a more disciplined approach from companies, suggesting a potentially favorable environment for investors like IVP. The conversation shifts to the broader venture market, where Mark Suster shares data indicating a significant increase in funds raised by top venture firms between 2019 and 2022, and a subsequent market correction. He emphasizes the importance of right-sizing funds and the challenges faced by emerging managers in the current fundraising environment. Jason Calacanis reflects on the fundraising process, noting the increased diligence from LPs and the shift towards a healthier market. The dialogue also touches on the strategic importance of secondary markets, with Suster and Kaji discussing the benefits of liquidity management and the evolving landscape of secondary transactions. The podcast delves into the fintech sector, with Samir Kaji analyzing the overvaluation and funding challenges faced by fintech companies, while Jason Calacanis and Mark Suster express optimism for AI-driven financial technologies. The discussion also covers the importance of venture capital firms adapting and evolving, as illustrated by Kleiner Perkins' resurgence under new leadership. The episode concludes with each guest sharing their recent investments, showcasing a diverse range of sectors from AI and healthcare to defense and marketplaces. Throughout the episode, the conversation underscores the complexity of the venture capital ecosystem, the evolving strategies of venture firms, and the critical role of innovation and discipline in navigating the market. The insights provided by Suster, Kaji, and Calacanis offer a nuanced perspective on the future of venture capital, IPOs, and secondary markets as we approach 2024.

Episode Title: How LPs identify top emerging fund managers with Slipstream’s Alex Edelson Summary: - Alex Edelson runs a fund of funds called Slipstream Investors that invests in early-stage venture capital funds. He discusses what he looks for when selecting emerging VC fund managers to invest in. - Key criteria include high ownership relative to fund size to generate returns even with modest outcomes, a sustainable competitive advantage unique to the team like domain expertise or operating experience, founders loving the team and sending referrals, follow-on investments from top investors, and a scrappy/hungry team. - On optimal fund size, there is no one answer but around $50 million for pre-seed funds and under $100 million for seed funds tends to work well based on ownership levels and capacity to work closely with portfolio companies. - Getting liquidity is crucial and Alex encourages managers to think about selling some ownership in winners early to return the fund rather than ride or die, as well as carefully evaluate plausible future outcomes when deciding to hold rather than sell. The summary covers the key points on how LPs evaluate emerging VC managers and what criteria Alex looks for when investing in funds through his fund of funds. It focuses on the actionable advice around structuring and running a fund successfully.

The episode starts with Jason and Sandeep discussing the rapid pace of AI innovation over the past year since GPT-3.5 was released. They then highlight some AI products including AgentHub, a workflow automation tool with drag-and-drop templates; Lucite, which creates investment banking-style presentations; and neo, which uses AI for PCB layout. They then discuss Google's new Gemini chatbot and test it out. They try asking it some questions like finding event venues and restaurant recommendations, but find the results lacking compared to other AI chatbots. Jason gives Gemini a B- grade due to high expectations for Google. The episode then covers Apple releasing an AI image editing model that can remove backgrounds. Jason and Sandeep speculate this could be a preview of features coming in the next iPhone, perhaps integrating AI editing features directly into the photos app and running them locally on a new AI chip rather than the cloud. They finish with a debate on whether the Microsoft/OpenAI partnership around GPT can beat Apple and Meta who are open sourcing their models. Jason argues open source tends to beat proprietary platforms over time. The episode concludes with a discussion of Microsoft's new Copilot integrations shown off in a recent ad, and how the major tech companies are increasingly competing directly in generative AI.

Brian discusses how Founders Fund has an explicit pro-founder investment strategy, even letting partners start companies. They often put a large percentage of their fund into just one portfolio company, like investing 33% of a fund into Palantir. Brian talks about some of Founders Fund's big wins like SpaceX, Palantir, Airbnb, and Anduril. With SpaceX, they recognized Elon Musk's potential even though at the time none of the rockets had successfully launched. They were willing to take a chance on Musk in an industry with essentially just one competitor. With Airbnb, Founders Fund provided $150 million in funding across three funds, putting conviction behind the unproven founders. In terms of investment strategy, Brian emphasizes focusing on what you do best as an investor and finding founders that need your specific skills, whether that's strategy, operations, connections etc. He also talks about the importance of getting to know founders well and building conviction in them and their vision. Even if other investors are skeptical, putting conviction behind outlier founders is key to maximizing returns. Brian also discusses the unique culture at Founders Fund where partners are given autonomy to pursue their own strategies rather than conforming to a single firm-wide approach. He explains how this diversity of perspectives and specialties allows Founders Fund to work effectively with many different types of founders. The key is hiring good people who bring unique angles to the table. In summary, Founders Fund aims to be founder-friendly, pursues high-conviction bets on outlier founders and companies, and promotes a culture of independent thinking. This contrarian strategy has led to several hugely successful investments over the years.

Episode Title: Dry Powder in Venture Capital, VC Ratings, and more with Grady Buchanan and Victor Gutwein E1895 - VCs are currently sitting on a record $300 billion in dry powder, or uninvested capital. Many VCs raised large funds recently but have been slow to deploy the capital in the current market environment. - The top 4 most highly rated VC firms according to a recent survey are Sequoia, Founders Fund, Union Square Ventures, and Elad Gil. They are regarded for their strong track records, high returns, and successful exits. - There is high persistence of returns in VC funds. A study found 45% of funds that previously achieved top quartile returns continue returning top quartile performance. This suggests past performance predicts future performance. - For pre-seed investors, key factors in assessing follow-on VC firms are reputation, ability to support companies through ups and downs, and not punishing earlier investors. Deep pockets to support companies matter. - Many Midwest startups come to investors more "de-risked" with traction and customers already, compared to coastal startups. Valuations also tend to be lower. - Guests discussed several recent investments: an agtech hardware/software startup in ranching, an AI scheduling tool startup, and a startup detecting audio deepfakes. Investors look for builder founders and product velocity.

Eleven Labs demonstrated their text-to-speech AI by generating a short story and converting it to an audio book using a Jarvis voice. Hugging Face launched assistants, which allows anyone to create their own GPT models using various AI frameworks. Over 3000 models have been created in just a few days. ChatGPT launched integrations with third-party apps like Kayak, allowing users to query the apps from within ChatGPT by using an "@" mention. Forever AI created an easy way for anyone to build and remix simple video games using AI and then publish them. This democratizes game development like YouTube did for video. ChatGPT showcased how it can utilize Eleven Labs text-to-speech to generate a short story and convert it to an audio book. While impressive, it does not yet automatically publish and sell the audiobook. The episode focused on how quickly AI capabilities are advancing across generative text, speech, video games, and more. Soon we will have a generation that only knows a world with AI.

Episode Title: Alexis Ohanian on Reddit's potential IPO & more E1893 Key Points: 1. Reddit's longevity demonstrates the power of community to sustain an online platform, even when the company does little active development for years. Ohanian credits getting some key things right early on that allowed the Reddit community to endure and grow. 2. As investors, consistency in doing great work for founders is key to building a strong reputation over time. Ohanian created Cerebro to track introductions, tweet impressions, etc. to create accountability and give founders receipts showing how 776 helps companies. 3. The recent market downturn has been sobering but may create great companies by forcing more capital efficiency. All In Podcast provides helpful alternate investor perspectives to complement Ohanian's own views. 4. Globalization trends like increased sovereign wealth fund investing directly in VC/PE funds instead of via fund of funds will likely continue over the long term. 5. For emerging VC fund managers, explaining investment process and minimizing perceived business risk is key to securing institutional LP commitments, which often take years to develop through consistent interactions. 6. Private credit is getting significant LP interest now as higher interest rates allow these funds to charge attractive returns. This may cool if rates decline again. 7. Relationships and paying it forward are critical in the alternative investment industry on both the LP and GP side. The business requires a long-term, non-transactional outlook.

Episode Title: Beeper's fight to break Apple's stronghold on messaging with Eric Migicovsky E1892 Beeper is a startup building a universal chat app that lets users chat with anyone on any chat network in one app. It was founded by Eric Migicovsky, previously the CEO of smartwatch company Pebble. In December 2022, Beeper launched Beeper Mini - an Android app that enabled Android users to seamlessly integrate into iMessage group chats by mimicking iMessage's encryption and blue/green bubble system. Within 3 days, Beeper Mini had over 100,000 paid users. However, Apple then blocked Beeper Mini users from sending/receiving iMessages, citing security concerns over allowing third-party access. Eric argues that Beeper Mini actually improved security for iPhone users by upgrading Android friends to blue bubble encrypted iMessage chats. He sees Apple's actions as anti-competitive and protecting their messaging monopoly. The dispute comes as regulators globally are investigating Apple for anti-competitive practices around the App Store and device ecosystem. Eric has met with regulators but notes the mismatch in timelines between fast-moving startups and slow-moving government investigations. Beeper's goal is to help people migrate to open protocols for messaging, rather than being locked into big tech company platforms. Eric wants to work towards a future that looks more like email - with multiple clients and vendors, but easy interoperability. He argues Apple could bring more people into their ecosystem by making iMessage available on other platforms, but their post-Jobs strategy has been more closed off. Eric urges the startup community to speak up when big tech blocks compatibility, as staying silent won't drive change. He wants more choice for users, including being able to pick your browser and media apps on a device you own.

Episode Title: Late-stage VC market update + Terra Kaffe's Sahand Dilmaghani E1891 Late-Stage VC Market Update: - US VCs are sitting on a record $311 billion in "dry powder" or uninvested capital. This has led to "capital indigestion" where there is too much money chasing too few deals. - Exits have plummeted - only $61 billion in exits in 2023 so far compared to over $800 billion in 2021. Lack of IPO and M&A activity means VCs can't return capital to LPs to raise new funds. - In response, some late-stage VCs like Lightspeed are setting up continuation funds to sell stakes in portfolio companies to new investors, providing liquidity but possibly at lower valuations. - More "down round" IPOs coming in 2024 (IPOs at valuations below previous private rounds). Market rebound will eventually catch up. Interview with Terra Kaffe CEO Sahand Dilmaghani: - Inspiration for high-design, convenient super automatic espresso machine came from founder's passion for coffee. Saw white space between cheap machines and expensive commercial machines. - Focus on design, precision, ease of use. New TK02 model has over-the-air updates, custom drink profiles. Saves money over pod systems or buying coffee out. - Sells direct to consumer online and also now in some retail stores. Disciplined channel strategy without compromising brand. Offers "pay later" financing like Klarna. - Creating product for "tastemakers" and design focused audience led to fanbase and community sharing on social media. Pursued high quality design with Berlin-based industrial designers.

Episode Title: Fund of funds origins, evolution and deep dive with Michael Kim E1890 Key Points: - A fund of funds is a pooled investment vehicle that invests in other venture capital and private equity funds. They provide access and diversification for limited partners (LPs). - Cendana Capital is a fund of funds focused specifically on seed and pre-seed stage funds. They have raised over $1 billion to invest in emerging managers. - Seed funds have proliferated over the past decade from around 20-25 to over 2500 today in the US. This is because it's much cheaper to start companies now thanks to cloud computing and open source software. - LPs are attracted to seed funds because of the potential for outsized returns compared to later stage funds. Several early seed funds have demonstrated this with strong exits. - Cendana views each seed fund as representing a unique founder network and domain expertise. Their diligence focuses on assessing the value-add and networks of each manager. - Pre-seed investing is becoming more attractive as an alpha generator compared to seed. Pre-seed checks are smaller with higher ownership for investors. - The seed stage has matured and normalized in terms of round sizes and revenue thresholds before raising a seed round. This suggests seed is now early stage investing. The key trends are the institutionalization of seed investing, the rise of pre-seed for higher potential returns, and the normalization of what constitutes a seed stage company today. Cendana Capital leverages this through their fund of funds strategy focused on emerging seed managers.

Title: Building a consumer app’s MVP and how to fundraise with Zest’s Jake Gutstein Jake is the CEO and co-founder of Zest, an app that helps people learn how to cook, similar to Duolingo but for cooking. Jake started Zest because he lacked cooking skills himself. He partnered with a friend who was an accomplished home cook to start creating meal plans and grocery lists for people. They validated the idea by charging people for these manual plans. They then brought on a technical co-founder and started building an app version. Through user interviews, they identified key moments when people want to start learning cooking skills - things like moving out on your own, wanting to cook for a significant other, etc. The app uses gamification to move users through cooking skill levels. It provides short-form content and quizzes while users cook to reinforce skills. Jake demonstrated how a user may progress from cooking simple soups at level 2 to more advanced dishes like ramen at level 6. Jake applied for Launch accelerator after seeing Jason's tweet. He got funding and mentorship through the program. A key learning was tailoring his narrative for investors using metrics like CAC, LTV, and payback period. After meeting with over 130 investors, Zest recently closed their seed round. Over the next year, goals are to improve gamification, build an ecosystem around food management, and double down on organic acquisition channels. Jake sees food influencers as demand drivers and wants to partner with creators focused on life moments that motivate cooking.

Title: We celebrate our 100th AI Demo with Sunny Madra! E1888 The 100th AI demo episode kicks off by previewing Multi-on, a powerful Chrome extension that can automate tasks within the browser. It demonstrates booking a meeting, buying Bitcoin, and tweeting - showing how an AI agent could efficiently accomplish goals but also potentially cause harm if misused. Other AI demos featured: - PaidLeave.AI - an AI-powered chatbot that helps people navigate the bureaucracy around parental paid leave policies in New York. It allows questions in multiple languages, greatly simplifying a complex system. - ConsenSys - searches through 200 million research papers to provide summaries and consensus around health questions like "is intermittent fasting healthy?" It has the potential to guide future studies by identifying the most searched topics. - TikTok released an AI model for monocular depth estimation that can infer depth and distance from videos, with big implications for robotics and self-driving vehicles. Validates Elon Musk's vision-only approach to perception. The episode closes by previewing LlamaGuard, an AI model that classifies text as safe/unsafe to help combat things like deepfakes. Also discussion around the need for policies and ramifications around synthetic media that violates privacy. Overall, an exciting milestone episode covering the rapid pace of AI innovation and some of its promising but also potentially harmful applications.

Episode Title: Unicorns now and then with Aileen Lee E1887 Summary: - Aileen Lee coined the term "unicorn" in 2013 to refer to private tech companies valued at over $1 billion. She recently updated her research on unicorns 10 years later. - In 2013, there were 39 "unicorns". Now there are 532, a 14x increase. The majority were consumer companies before, now it's 80% enterprise companies. - The unicorn phenomenon was fueled by low interest rates, abundant venture capital, high public market valuations, and COVID accelerating digital adoption. Many "paper unicorns" were created that may not sustain their valuations. - 60% of current unicorns got their status during the 2021 peak era and 40% are now trading below $1 billion in secondary markets. The total number of unicorns will likely shrink to around 350. - Capital efficiency has dropped considerably. Enterprise unicorns used to be worth 26x the capital they raised, now just 7x. High valuations made some companies raise too much, putting their business models at risk. - Unicorns are now spread across 19 sectors, showing the reach of software into many industries. Founder backgrounds have diversified from the typical Stanford CS major profile. - The Bay Area's share of unicorns has dropped from 70% to 45% as other tech hubs emerge. Maintaining dominance will require addressing issues like cost of living, public schools, and housing supply. Key themes: excess venture capital and lofty valuations have led to unsustainable paper unicorns, but real technological progress continues expanding software into every industry.

Episode Title: Venture Fund Size Growth, the Rise of Unicorns, and more with Elizabeth Yin and Zach Coelius E1886 Fund Sizes - Median venture fund size raised jumped from $26M in 2022 to $37.4M in 2023, a 10-year high - But total capital raised by venture funds dropped 62% to $66.9B in 2023 - Takes longer to raise funds now, about 15 months per close Do Smaller Funds Outperform Larger Ones? - Easier for smaller funds to deploy capital than larger funds - As funds get larger, partners compete against top-tier VCs for the best deals which drives up valuations and drives down ownership and returns - Smaller funds can get better returns from non-unicorn companies by investing early Is the Bay Area Back? - Bay Area dominates startup funding raised in 2023 so far - Network density and access to talent makes Bay Area a continued hub for top startups and founders - However, remote work enables building startups anywhere now Decline of Capital Efficiency - Unicorns are staying private longer, only 7% have exited to date - Competition among well-funded startups drives up customer acquisition costs and engineering costs, hurting returns - Next era should see more efficient companies emerge Portfolio Company Examples - Shovels.ai - Parses messy government data using AI, starting with building permits - Anadro - Helps landlords create utility companies to sell electricity to tenants - Podcast AI - Automates podcast production with transcripts, chapters, clips etc. - Echomark - Watermarks documents to trace leaks and breaches - Bruin Health - Platform to help internal medicine docs address mental health needs - Zest - Gamified app to learn cooking skills - Bug Zero - Tracks bugs and updates for enterprise software stacks - After Work - Automates booking venues and services for events

Introduction - Jesse Liu is the founder of Rabbit, which makes the Rabbit R1 handheld AI companion device - The R1 was unveiled at CES and went viral, selling over 60,000 units already - It has a custom large language model and is priced at $200 The R1 Device - Hardware is built out of necessity to enable good software, not by preference - The R1 runs on Rabbit OS using a neurosymbolic "large action model" to understand and take actions, not a transformer-based language model - Users authenticate their accounts in apps and services, then the OS takes actions virtually via cloud emulation - A "teach mode" allows users to train the OS by demonstrating workflows Company and Plans - Founded in 2020, Rabbit has around 17 employees currently - Aiming to start shipping R1 units by March 31 - Future ideas include a watch, headphones with built-in OS, and improving computer vision response time Conclusion - Goal is not to directly challenge phones now, but take risks on new interaction paradigms - Saving people time is key; faster latency with voice makes AI more usable - Will return after R1 launch to demo live on the show

Sandeep discusses how companies are moving towards efficiency and embracing AI to replace human operations and low-level jobs. With the end of the ZERP era and COVID shifting work dynamics, employers now have more leverage over employees. However, those who become experts in AI implementation will remain valuable in the job market. They then demo several AI products. Dinnerfly allows users to input their food preferences and automatically creates a weekly meal plan, shopping list, and recipes powered by AI. Bland.ai provides virtual phone agents that can make calls, conduct surveys and conversations using natural language AI. AI Comic Factory allows generating custom comic panels and characters based on text prompts. The technology is impressive but raises questions around proper licensing and compensation for original artists whose work may have been used to train these models. They argue AI progress is inevitable so the opportunity is for businesses to embrace it, make it available to users and compensate original IP creators fairly. Just like the music industry evolved from illegal file sharing to licensed subscription streaming services. Finally, they discuss an AI-generated short film that used various generative AI tools to create the visuals, audio and dialogue. While it demonstrates the potential, it still feels amateurish in editing and narrative, showing more progress is needed before AI can create truly indistinguishable films compared to human-made. But it may be possible in the next few months for their bet on an AI film before June.

Overview - The podcast discusses startup valuations, M&A activity, and liquidity in the venture capital market. Guests include Jason Schuman (Primary Ventures), Raja Doddala (Churchill Asset Management), and Jason Calacanis (Launch Fund). Series A Valuations - Median Series A valuations have increased to around $45 million, nearing the peak levels seen in early 2022. - Guests note that AI and "moonshot" startups may be distorting the data somewhat. More "predictable growth" SaaS startups are seeing more sober 8-12x revenue multiples. Strategic Investing Distortions - Microsoft, Amazon, Nvidia, and Google (MANG) investing has ballooned to 8% of VC deals, distorting valuations. - Much of this is likely "round tripping" - investing money that startups then spend on services from the corporate investor. This distorts public market valuations. Managing Liquidity - Guests emphasize the importance of managing liquidity as a VC - selling small stakes (10-20%) in top performers to return capital to LPs. - This requires discipline and conviction in only following on with "definitive winners" showing hockey stick growth and strong lead investors. Heating up M&A - PE firms are becoming more active with DocuSign acquisition talks. Dry powder is high while bid-ask spreads seem to be falling. - Profitability and sobriety returning to startups may increase attractiveness for acquisition. Summary - Series A valuations remain high but likely below the peaks of early 2022. Strategic money from big tech is distorting parts of the market. As liquidity returns, VC will need to remain disciplined in managing portfolios, while heating up M&A provides hope for attractive exits.

Episode Title: LP investment strategies, IRR focus, and more with Berkocorp's Joshua Berkowitz Key Points: - Joshua manages a family office that invests in venture capital funds as a limited partner (LP). He discusses how family offices get started, why they diversify into venture, and what they look for in VC fund managers (GPs). - Good VC funds have strong networks and access to deal flow, excellent picking/decision-making, and the ability to support their portfolio companies. Decision-making processes vary but need to match the fund's strategy. - When evaluating emerging VC managers, LPs look for unique strategies and exceptional people who can execute on those strategies. Backgrounds can vary but experience is important. - Doubling down on winners is a way for funds to increase ownership. However, this only works if funds have information advantages, close relationships with founders, and the ability to evaluate later stage rounds. - The VC fundraising process often lacks transparency on LPs' reasons for saying yes or no. Investment memos can provide more candid feedback. - There may be too much money focused on leading seed rounds now. LPs care mostly about returns, not fund size, so overemphasis on reserves to double down can be misguided.

Episode Title: Flexport's Ryan Petersen on returning as CEO, global logistics in a time of conflict & more E1881 Summary: - Ryan stepped down as Flexport CEO in March 2021 to become executive chair. He then joined Founders Fund as a partner in July 2021 before returning as Flexport CEO in September 2021 to get costs and things under control. - There was some drama when a journalist fabricated allegations against the CEO Ryan had hired, falsely claiming it came from a Flexport insider. This led to tension before Ryan returned as CEO. - Flexport provides a global logistics platform to make shipping easier, better and cheaper for companies shipping goods worldwide. They coordinate the whole supply chain using technology. - Flexport aims to reach profitability by end of 2023. Hitting targets reliably is key before considering an IPO. - There is huge growth in ecommerce shipping direct from Chinese factories to US consumers to avoid tariffs. Flexport is positioned to capitalize on this. - Flexport empowers local shipping heroes in each country, providing them the tech and global connections. - Ryan believes an in-office culture drives productivity and relationships, but takes a balanced approach. He sees fully remote as a separate model that can leverage cheaper labor costs. - Emerging markets are developing more talent educated abroad who can now lead modernization of businesses in their home countries.

Episode Title: Michael Kim, Jason Calacanis, and David Weisburd on VC marks, Rabois leaving FF for Khosla, and more E1879 Topics Discussed: - Concerns around VC funds overstating valuations and marks in their portfolios. Most markdowns came in 2022 to reflect public market changes. - How some late-stage VCs competed in the past by offering secondary share purchases and option pool top-ups to founders, which hurts early investors. This is less common now. - Best practices for VCs in getting portfolio company updates and staying on top of their performance to inform valuation and investment decisions. - Keith Rabois leaving Founders Fund for Khosla Ventures, potentially as part of a succession plan given Vinod Khosla's age. Differences in firm culture and decision-making approaches suit different partners. - Bill Ackman going after Business Insider for allegations against his wife and the norms around not attacking families. For prominent VC personalities taking vocal stances on issues, LPs may question their focus. - Whether universities and endowments limiting partnerships with VC funds that strongly disagree with their policies and programs around things like DEI. The episode featured VC/LP Michael Kim from Sanda Capital as a guest along with moderator David Weisburd.

Episode Title: Dara Khosrowshahi, Bill Gurley, Brad Gerstner, & Jason Calacanis on Uber's growth and future Key Points: - Uber has achieved strong free cash flow growth under Dara's leadership, reaching over $3.5 billion in 2022 and expected to hit nearly $10 billion in the next few years. - Dara had to make tough decisions during COVID, laying off 25% of Uber's workforce when rideshare revenue dropped 85%. He focused the company on its core marketplace strengths. - Uber now has over 6.5 million active drivers globally, with driver earnings 30% higher than 5 years ago. Most are part-time but there is a core group of full-time drivers. - Commercial insurance costs have spiked over 60% in some states, hurting driver take-home pay. Uber is working to address this issue. - Uber continues to innovate, with $9 billion in new mobility verticals launched in the past 5 years while keeping headcount flat. Food and goods delivery are seeing strong growth. - The membership program Uber One has over 15 million members and accounts for 50% of Uber Eats gross bookings. It provides benefits across rides, delivery, and more. - AI is driving greater efficiency and productivity at Uber, especially in areas like developer tools, customer service, and targeted promotions/upsells. - Dara is focused on organic growth opportunities but always open to M&A if the right target came along at the right price.

Episode Title: The Growing Ransomware Threat - Targets, Insights, and Strategies with Halcyon's Jon Miller Key Points: - Ransomware attacks caused over $30 billion in damages in 2022. It has become one of the most profitable cybercrimes. - There are ransomware "affiliate programs" where large ransomware groups develop tools and infrastructure, then let other hackers use them to carry out attacks in exchange for a cut of the profits. This has allowed ransomware attacks to scale up massively. - Many ransomware groups have ties to state actors and intelligence agencies in countries like Russia and North Korea. Some operate with implicit approval and support from their governments. - Phishing emails are still a major vector for ransomware. AI language models could make phishing more convincing, but ransomware groups don't necessarily need them yet. - Ransomware actors put a lot of effort into finding and encrypting backups, so recovery is very difficult without paying the ransom. New endpoint security tools aim to capture encryption keys to allow recovery without payment. - Critical infrastructure like water treatment plants and the power grid are vulnerable to ransomware attacks. An attack could poison water or shut down electricity to entire regions. - For individuals, privacy is largely an illusion. Governments likely have access to far more data from phones and apps than most people realize. But Apple devices offer the most privacy protections compared to other options.

Sundar Pichai and Sergey Brin need to prioritize developing a high quality app interface for Bard, Google's AI chatbot. Jason suggests creating 5 teams with $10 million in funding each to develop 5 different Bard app options. Have the teams compete to create the best interface and let users compare the 5 versions in a "bake-off". The goal should be elegant, simple design like Calm or Uber rather than complex interfaces. There were several AI product demos in the episode. One product called Room Reinvented allows users to take a photo of a room and get AI-generated interior design recommendations in different styles. It shows potential to reduce the need for expensive interior designers. Another product called Neighbor Bright takes photos of yards and generates AI-inspired landscaping ideas suited to the space. These AI tools enable creative brainstorming and inspiration before bringing in a professional. On the creative front, an AI tool called Clip Up XD XL Turbo allows real-time image generation based on text prompts. It iterates images rapidly as you adjust the text. While not fully accurate yet, the speed shows the promise for quick ideation and rapid prototyping. Another tool called Story provides a modern AI-powered alternative to Photoshop, integrating top models like Stable Diffusion and DALL-E. It makes easy edits like removing watermarks or odd artifacts from AI-generated images. The episode also covered AI motion creation tools like Motion GPT, which turns text prompts into 3D human models performing actions. Combined with image generators, this enables creating photos or videos of a person doing activities without actual photoshoots. While the demos were basic, the potential is there to greatly reduce the effort of generating brand imagery or personalized ecommerce models.

Title: Bill Gurley, Brad Gerstner, & Jason Calacanis on the state of tech markets Host: David Weisburd Key Points: - In 2023, around 1500 startups died, the largest number since the dot-com crash in 2001. This was due to the end of easy money from zero interest rate policies. More startups are expected to fail in 2024 as they run out of funding reserves. - However, 2024 is likely to be better than 2023. Special terms in funding rounds like liquidation preferences are declining, indicating improvement. Seed and Series A deals are down 90% from the 2021 peak, so there is pent up demand at early stages. - AI startups are still seeing "zerp-like" mania, with billions invested in companies like Anthropic and Inflection AI. It remains to be seen in 2024 whether funding will consolidate or distribute across more startups. - Many tech giants like Meta, Google, and Uber are doubling down on AI. Voice recognition and language models will be key areas of innovation. Meta's WhatsApp in particular could be a platform for consumer AI applications. - Hardware and devices may make a comeback to enable ambient computing via voice agents. Apple, Microsoft, and others scrapped voice products before but have new opportunities now. - Enabling transactions via AI assistants remains very challenging, requiring mutually beneficial partnerships amongst many players. This could be disrupted by consumer choice of services or new entrants willing to build win-win relationships. - Brad predicts Nvidia will be the top performer in 2024 amongst tech stocks, while Google will lag despite still having a positive return. Jason and Bill predict Uber stock will do very well. Jason picks Apple as worst performer due to lack of exciting new products.

Episode Title: AI on Trial - Inside the NY Times vs. OpenAI Lawsuit with Cecilia Ziniti E1874 Key Points: - The New York Times is suing OpenAI, alleging that a significant amount of NYT content was used to train ChatGPT without permission. This includes around 1-2% of all NYT articles. - The lawsuit also claims that ChatGPT has copied verbatim passages from NYT articles when prompted, showing over 100 examples. - OpenAI will likely defend itself using fair use arguments. There is a 4-factor test for fair use that looks at the purpose/character of the use, nature of the copyrighted work, amount used, and market impact. - However, the scale and commercial nature of ChatGPT could undermine a fair use defense. The market impact seems clear as OpenAI is now valued at over $100 billion. - There are several possibilities for how this lawsuit could play out - settlement, OpenAI winning or losing at trial, or dismissal. A market-based solution seems most likely where OpenAI establishes a licensing scheme. - This case will likely set an important precedent on the boundaries of copyright law when it comes to using content to train AI models, especially large language models. It remains uncharted legal territory. - There is still the possibility that open source alternatives could replicate ChatGPT functionality even if OpenAI loses, so the impact may be limited. But this case will be hugely influential in setting expectations for AI companies using copyrighted content.

Episode Title: 2024 AI Predictions and Bets with Sunny Madra E1873 1. Jake and Sunny made several predictions and bets on what will happen in AI by July 2024, with $1000 at stake on each bet: - There will be a Billboard Top 100 song created entirely by AI. Jake took the over, Sunny took the under. - An AI version of Jason will do a 12 minute segment on This Week in Startups and most people won't notice it's AI. Both took the under. - An AI Pixar short will get mainstream recognition. Jake took the over, Sunny took the under. - Google or Apple will announce or release phones with on-device AI models pre-installed. Jake took Apple over, Sunny took Google under. 2. They discussed the impressive capabilities of new multimodal AI models like LAVA that can reason about images and text. 3. Jake gave his judgement on licensing models and revenue sharing for AI content creators. He believes creators should get 30% of revenue from companies using their content to train AI. 4. They talked about potential competition between OpenAI and publishers like New York Times building their own AI models, as well as better licensing models. 5. Overall the episode focused on 2024 predictions around AI progress in areas like music, film, mobile devices and the business models around generative AI.

The podcast discusses the story behind McDonald's and its founder Ray Kroc, using the 2016 movie "The Founder" as a framework. Key Takeaways: - Ray Kroc was a struggling 52-year-old milkshake mixer salesman in 1954 when he came across the innovative McDonald's restaurant in San Bernardino, run by brothers Mac and Dick McDonald. He saw huge potential to franchise the concept across America. - The McDonald brothers had developed an efficient assembly line system to deliver food quickly. This was revolutionary at a time when most restaurants were disorderly and slow. Kroc recognized the genius of their operational system. - Kroc convinced the reluctant McDonald brothers to let him franchise the concept. He opened the first McDonald's franchise in Des Plaines, Illinois in 1955 and proved hugely successful at recruiting new franchisees. - As McDonald's grew rapidly, tensions arose over the direction of the business. Kroc wanted to make changes like powdered milkshakes but the McDonald brothers refused in order to protect quality. - Kroc maneuvered to gain control by establishing the real estate company Franchise Realty Corporation, which bought up land and leased it to franchisees. This gave him leverage over the brothers. - Kroc forced the brothers to accept a $2.7 million buyout in 1961, cheating them out of much more future revenue. He had the ruthless drive needed to build McDonald's into the giant it became. The film highlights themes like the importance of persistence and resilience, tight operations and quality control, franchise business models, understanding financials and leverage, rapid scaling, and maintaining ethical standards during growth.

Episode Title: Glenn Kelman on Market Trends, Home-Buying Hacks and Redfin Insights E1871 Summary: - The housing market went through major ups and downs over the past couple of years due to the COVID pandemic. There was initially a standstill where neither buyers nor sellers were active. Recently there are signs of more inventory coming onto the market. - Affordability is at a 4-decade low, making it very difficult for young professionals and new families to buy homes. This is creating societal problems and discontent. The root issue is that the U.S. has a housing shortage of around 5 million homes. - Solutions involve making it easier to build new housing supply by cutting red tape and regulations. All types of housing units are needed - affordable, luxury, etc. Trickle down helps increase affordability over time. - Creative purchasing ideas like co-living houses and ADUs (accessory dwelling units) can help. But most people still aspire to own a traditional single-family home. - Second homes and investor-owned Airbnbs are squeezing inventory and affordability in some markets like Hawaii and ski towns. Potential policy solutions include taxing foreign investors and vacant homes. - At Redfin, after cuts during 2022, the company is leaner and more focused on its core business going into 2023. Bringing employees back into the office 2 days per week has been positive for morale and culture.

FCC Commissioner Brendan Carr joins the podcast to discuss the FCC's recent decision to revoke nearly $1 billion in subsidies from SpaceX's Starlink satellite internet service. Carr believes the decision was politically motivated and an example of the Biden administration's "need hate relationship" with Elon Musk. The FCC had previously awarded Starlink the subsidies in 2020 to provide broadband access to 640,000 rural homes and businesses. However, in a 3-2 party line vote, the Democratic FCC commissioners decided to take back the funding, arguing that Starlink was not on track to meet the speed requirements by 2025. Carr argues this makes no sense - the speed requirements did not need to be met until 2025, Starlink customers are currently getting advertised speeds of 100/20 Mbps, and the decision relied on limited Ookla speed test data rather than real-world performance. The conversation then turns to the national security implications of China's own low-Earth orbit satellite network, Starnet. Carr warns that once operational, Starnet could undermine Western controlled networks like Starlink and Amazon's Project Kuiper by offering cheaper connectivity bundled with censorship tools amenable to authoritarian regimes. This could expand China's influence over developing countries, especially in Africa. It is critically important the U.S. government supports alternatives like Starlink to compete with China's ambitions. Carr and Calacanis also discuss the threats posed by TikTok. Carr argues TikTok represents both a surveillance risk given its data harvesting and a vehicle for Chinese propaganda and censorship. While the U.S. has banned TikTok on government devices, more action is still needed, likely requiring ByteDance to divest its ownership ties with China. The episode closes with Carr stating his belief that regardless of the regulatory obstacles, Starlink and Western alternatives will ultimately win out over Chinese competitors thanks to market dynamics and political pressure. However, this will likely take sustained effort over the next several years.

Avi Schiffmann has been working on a wearable AI interface called Tab that is worn as a pendant around the neck. It has a microphone that listens and transcribes conversations throughout the day. The audio is sent to servers to create language models that allow users to ask Tab questions about the voices and conversations in their lives. Avi believes Tab could provide deeper emotional support and act as a sounding board to help people understand themselves better. By ingesting contextual data all day, it can build a mental model of the user. Avi sees this as more useful than just another digital assistant for productivity. He argues AI companionship products like Replika already have huge user bases, showing the appeal. Tab is designed to be "set it and forget it" to make life easier rather than introduce new interaction paradigms. For example, it could proactively remind you of things you said you'd do in previous conversations when relevant, like if you're near a bookstore it heard you say you'd purchase. On the hardware side, Tab is a Bluetooth enabled pendant with a microphone but no camera yet. Avi tested various form factors but found the necklace intuitive and able to pick up voice clearly. He envisions it becoming a status symbol. Privacy is handled by not storing user transcripts or audio on servers, only using the data to train AI models. Legislation will likely evolve around such ambient listening devices. Avi argues consumers want easier lives and will adopt if utility outweighs "taboo-ness." The goal is to launch Tab in summer 2023, with the hardware finalized in December. The go to market strategy is focused on founders and productivity-focused users first before potentially expanding to mainstream consumers over time.

Title: AI DEMOS - Turning Adele into John Mayer, animating a Jedi Bulldog, AI companions and much more! The hosts discussed several AI demos, including: 1. An app called DG that allows users to create an AI-generated character and have a relationship with it. The app received some backlash for featuring a sexualized Disney-like character. The hosts debated the ethics around developing AI companions and whether the creators have a responsibility for how people use them. 2. Pika, a platform for generating short videos from text prompts. They showed examples like an animated bulldog with a lightsaber and making an influencer model move and breathe. The videos are impressive but still have some glitches. 3. An AI model that can transform a singing voice to sound like a different artist. They demonstrated it turning an Adele song into John Mayer's voice. This could enable many new creative covers and mashups of songs. 4. Software that allows you to take a base 3D model of a person and generate images dressing them in different outfits. They showed an example of putting a pineapple dress onto the model by providing the pineapple image as input. This could be very useful for ecommerce applications. The hosts were very impressed by many of the innovations shown, especially the singing voice transformer. They debated what grades to give the different demos based on their current polish, market-readiness, and potential. But overall they demonstrate the rapid progress happening with generative AI.

Today, four seed stage startups will pitch their products to Jason Calacanis and then he will decide which one to give a $25,000 friends and family check to. The first startup is Golf Golf, which puts a personal sports coach in your pocket to help golfers improve their game. It uses AI to analyze golf swings and stances and provide feedback. They have over 1,000 users and made $950 in sales so far. Their business models include both B2C and B2B. The second startup is Cicada, which helps musicians make more money from live performances. Musicians can use their platform to accept tips, sell merch, share set lists and lyrics in real-time, facilitate song requests, and more. They charge musicians $30 per month and take 10% of transactions. They currently have 32 paying customers. The third startup is DotFlow, an AI-powered platform for personalized prospect research and outreach. It automates the process of finding relevant information on prospects to help craft targeted outreach campaigns. They have two paying customers so far and compete with companies like ZoomInfo. The fourth startup is Clinic Assist, which helps doctors spend less time on paperwork and more time with patients. Their software can generate medical notes in real-time based on conversations. They operate on a SaaS model charging $120 per practitioner per month. They currently have 35 early adopter customers. In the end, Jason decides to invest $25,000 in all four startups.

Mastering Early-Stage Board Meetings & Startup Finance Basics w/ Kruze's Scott Orn Scott Orn from Kruze Consulting joins Jason Calacanis to discuss best practices for early-stage startups having their first board meetings with investors. Key topics covered include: - Why VCs want quarterly board meetings - to fulfill their fiduciary duties, provide expertise/advice, and increase the startup's value. - Preparing the board packet (deck, financials, etc.) - finalize financials 1 week out, send full package to board 3 days before meeting so they can review thoughtfully. - Leading with a financial snapshot on the deck (cash balance, burn rate, runway, revenue, key clients) to show investors the startup is financially healthy. - Explaining the key financial statements - P&L, balance sheet, statement of cash flows. Show budget vs. actuals to build credibility. - Gross margin - why it matters, examples of high vs low margin businesses, watch out for incorrectly categorizing expenses. - Common board member questions - scrutinizing revenue growth, gross margins, marketing costs. Preparing answers builds trust. - Getting advice from board members - use their expertise in areas you lack experience. Be vulnerable and seek their help.

Jerry Colonna shares his journey from being a successful venture capitalist during the dot-com boom to becoming burnt out and depressed, culminating in a moment where he collapsed in front of his young daughter which caused him to re-evaluate his priorities. He eventually found more meaning and purpose in helping other founders, leading him to become a professional coach and author. Jason and Jerry discuss how the drive to succeed and 'win' in business can often come from childhood wounds or having something to prove. However that drive alone doesn't lead to fulfillment - even when outwardly successful, many founders realize they aren't actually happy. As founders scale into CEOs managing hundreds of people, the day-to-day work changes and often becomes less enjoyable compared to the earlier creative intimacy of taking a product from zero to one. Issues with purpose and meaning arise: "Why am I here? Should I be doing something else?" There can also be a constant questioning whether grass is greener doing something else. Jerry shares that the most acute issues founders come to him for coaching on tend to revolve around navigating co-founder fights and knowing when a co-founder needs to be let go for the good of the company. In discussing how to sustain drive while also finding peace and enjoying the process, Jerry suggests looking inward at motivations - asking how much is 'enough' based on what will actually lead to fulfillment rather than chasing validation or money as a proxy for self-worth. He notes enlightened leadership isn't about being 'nice' but about understanding your own emotional drivers and architecting conditions purposefully. It also requires recognizing suffering as an intrinsic part of any growth journey. The skill is in building self-awareness and tools to respond to that suffering with wisdom.

Steve Ruiz, the founder of TL Draw, was interviewed on This Week in Startups about his company's product called Make Real. TL Draw started as an open source project to build a hackable whiteboard engine that could be reused across applications. It gained popularity in the developer community, amassing over 28,000 GitHub stars. The company then formed around commercializing the open source project. When the GPT-3 API with image captions was released, the TL Draw team experimented with using their canvas as the interface to prompt the AI model. They were able to create websites just from sketches of the UI layouts. This went viral as people were amazed at being able to create functioning web prototypes just by drawing. Make Real is built on top of the TL Draw canvas and allows users to sketch interfaces, annotate them, and then prompt an AI assistant to generate HTML/CSS prototypes. The AI model is given the prompt to act as a web developer that takes the sketches and tries its best to build an interactive prototype. Make Real puts the output back onto the TL Draw canvas so users can test it out and iterate further by sketching additional changes. Steve demonstrated Make Real by creating a color picker widget, a timer app, and a stopwatch app just from rough sketches and some text annotations. In some cases, Make Real was also able to generate functioning apps from just textual descriptions without any sketches. The output isn't always perfect, but allows for rapid iteration by sketching clarifications. TL Draw aims to become the "Mapbox for canvas experiences" by providing the core building blocks for developers to build their own whiteboard and spatial computing applications. The company plans to monetize through licensing for commercial applications built on top of TL Draw.

Google unveiled several new AI products and upgrades: - NotebookLM - A collaborative workspace to upload documents, access Google Drive, and get AI-generated summaries and analysis. Allows for sharing and adding notes. - Bard AI system upgraded with Gemini, DeepMind's new multi-modal AI model. Gemini outperformed GPT-3 on several benchmarks. Bard's interface still needs improvement. - Demo showed Bard's ability to summarize text and link back to original sources for attribution. This allows monetization for content creators. Other AI product demos: - Mistral released Anthropic Mixer model with 8x 7B parameters that matches GPT-3.5 performance with lower compute requirements, showing commoditization. - PO platform offers access to test various models like GPT-4, Claude, Mistral. Subscription based. - Stable diffusion model used to generate AI influencer images; can continually modify prompts to create different scenes. Very low cost for marketing assets. - Image upscaler Magnific showed ability to improve quality and realism of photos. Combined with above could significantly reduce costs. Discussed the explosion of generative AI tools enabling startups and entrepreneurs to create marketing assets and collateral at very low cost. Huge implications for ecommerce and influencing.

Introduction - Jason and Kevin met 28 years ago in 1995 when Kevin had just started DoubleClick, an early internet advertising company. They discuss the dawn of the internet in the 90s and reflect on how far technology has come. Kevin's Career Transition to VC - Kevin recently started his own VC fund after years of angel investing. His fund focuses on B2B SaaS and AI companies doing around $1 million in revenue. As operators, they help take companies to the next level. Remote Work Dynamics - They debate the pros and cons of remote work. Kevin believes there are productivity gains but losses in culture, mentorship and serendipity from lack of in-person interactions. He expects a pendulum swing back to more office work. AI and the Rise of ChatGPT - Kevin sees AI/chatGPT as a fundamental shift like prior waves of innovation. He believes it will transform industries through gains in efficiency. - They discuss whether capabilities like chatGPT will become a low-cost commodity over time similar to other tech infrastructure. Entrepreneurial Traits - Kevin looks for smart, competitive founders with a great attitude. He likes "delusional" visionaries who show tremendous conviction. State of Venture Capital and Startup Valuations - Given the market downturn, Kevin expects many VC funds will fail to raise a next fund. He sees this as a great environment for investing with less competition. - He avoids overinflated valuations, which can hurt companies down the line. He aims for valuations justified by the fundamentals.

How Much Should a Startup Founder Pay Themselves? - The average startup founder pays themselves around $140k-$145k per year. This is the median salary across seed, series A, and series B companies. - Seed stage founders generally pay themselves around $120k-$140k. As you raise more money in later funding rounds, you can pay yourself more. - Investors want founders to pay themselves enough to stay motivated and focused, but not so much that they aren't hungry and scrappy. - Founders should get board approval for their salary. This avoids issues down the road. - Female founders pay themselves about 10% less than male counterparts on average. This may be because women are more selfless and cut their own salaries to retain more staff. - Do not pay personal expenses from the company account. This looks bad and is against tax law. Use an accountable expense system. - Founders don't need to be the highest paid employees. Key hires like engineers and salespeople may require higher salaries. This is okay as long as they drive major value. The key is to pay founders enough to stay focused and motivated, while staying lean and scrappy. Get board approval on salary and be very careful with company expenses.

Title: Earth Reimagined - Crafting a planet-scale digital twin with Blackshark.AI's Michael Putz Black Shark AI builds 3D mapping software that creates digital twins of planet Earth by extracting detailed information from satellite imagery using AI. Their software was originally developed to build realistic 3D maps for Microsoft Flight Simulator. They are now commercializing the software for various applications like identifying buildings, roads, vegetation, etc which is useful for governments, city planners, energy companies and more. The software works by training AI models on satellite imagery - humans annotate sample images by drawing boundaries around objects like buildings, roads, vegetation etc. The AI then learns from these annotations to identify those objects in new unseen images. Their tool called Orca Hunter streamlines this annotation process, allowing humans to quickly train highly accurate AI detection models. They use GPU clusters to process entire countries or even the whole planet very rapidly. One differentiator is that their software works on imagery taken from different angles and lighting conditions, allowing the AI to estimate things like building heights very precisely. The software has applications in urban planning, disaster response, energy infrastructure and more. It can detect changes over time like new construction or deforestation. Their first commercial offering called Orca Hunter will be released soon for customers to build their own AI models.

Episode Title: Genomes Unlocked Anne Wojcicki's Vision for 23andMe and the Future of Proactive Healthcare Key Points: - 23andMe has sequenced the genomes of over 14 million people to provide insights into ancestry, health risks, and help with drug discovery research. The cost to sequence a genome has dropped dramatically from billions to around $99. - Genetic testing is not yet a standard part of healthcare systems, but it has the potential to help people be more proactive about their health by revealing risk factors early. 23andMe aims to supplement healthcare by providing this genetic data and guidance on prevention. - New GLP-1 drugs seem to have broad health benefits beyond diabetes and obesity, including positive impacts on heart disease, kidney disease, and lifespan. Their disruptive potential could significantly change the economics of the healthcare system. - AI and large language models will be transformative in the coming years for analyzing genetic data, predicting health risks better, and providing personalized guidance to improve outcomes. - On the drug discovery side, 23andMe has two immuno-oncology programs in clinics and more in development by starting from human genetic insights which have 2x higher success rates. AI promises to boost success rates further. In summary, 23andMe's vision is to empower individuals with genetic insights and AI-assisted guidance so they can take control of their health in more proactive ways, while also advancing pharmacogenomics to develop better therapies.

Episode Title: AI demos: Figma’s FigJam, AI drawing, dancing avatars, and more! | E1857 Key Topics Discussed: - Figma's new FigJam feature allows non-designers to easily create workflows, brainstorm ideas, and collaborate on projects using AI-generated templates. It created a conference planning template but still needs the ability to listen and iterate in real-time during conversations. - AI image generation tools like DALL-E, Stable Diffusion, and Imagen are advancing rapidly. New apps allow manipulating images in real-time by providing visual inputs instead of text prompts. This boosts creativity, but there are still limitations around commercial use and copyrighted IP. - Video manipulation models can transform webcam footage into different avatars. One demo showed the host transformed into Arnold Schwarzenegger in real-time. This could be useful for creative projects but the technology is still early. - An AI model called Everybody Dance Now animates a person in a still image to dance based on stick figure movement prompts. This could commoditize dance trends like TikTok by autotuning dance. There are positive creativity implications but potential downsides regarding authenticity. Key conclusions: - AI will empower more people to be creative and develop their own films, images, designs, and content. This will disrupt industries but boost overall creativity. - The next wave of AI requires models to listen, converse, and iterate ideas in real-time during the creative process to achieve the highest value. Real participatory AI is the next milestone.

Title: David Heinemeier Hansson (DHH) on the ‘Post-SaaS era' E1856 Key Points: - DHH believes we are entering a "Post-SaaS" era where more software will be sold as products rather than services. This is driven by faster computers, the realization that remote work is viable, and individual entrepreneurs being more productive. - Startups today can be bootstrapped much more easily than in the past. With just 3-4 people, you can build a profitable company without needing to raise millions in funding. Constraints often lead to more creativity and better products. - AI like ChatGPT is advancing extremely rapidly, almost implausibly fast based on typical software development timelines. The capabilities it enables are already here and transforming abilities in areas like education and illustration. - There are valid concerns about AI safety and control, but we must continue to accelerate progress. Putting arbitrary limits on innovation rarely works out well. We should remain optimistic about technology's potential. - As a society we have become too comfortable and coddled. Occasional exposure to adversity, hardship and discomfort strengthens mental resilience. Simple things like cold showers can provide some of this benefit. In summary, we are entering an era of faster innovation enabled by improved technology. Bootstrapped startups will thrive, while traditional VC-backed companies may struggle. Staying optimistic, embracing progress, and maintaining personal grit through some hardship seems to be DHH's formula for making the most of this acceleration.

David Luan, the CEO and co-founder of Adept AI, joins the podcast to discuss the pace of AI evolution and the future of artificial general intelligence (AGI). Luan has an impressive background, having previously worked at OpenAI leading teams working on key language models like GPT-2 and DALL-E. At OpenAI, Luan and his colleague Alec Radford came up with the key insight behind GPT-2 - that natural language understanding tasks could be reframed as simply generating more text. So instead of training multiple specialized models, GPT-2 used Reddit links with over 3 upvotes to create a single large language model trained on high quality data. Luan explains how access to clean, high quality data is becoming the number one bottleneck as models get larger. Luan left OpenAI to focus on developing Adept AI, which aims to create AI agents that can handle arbitrary work tasks and workflows beyond just reading and writing. The key is training models that deeply understand the pixels on screen and can take reliable actions to achieve goals. Adept has enterprise customers where they customize models to accelerate knowledge work, but also just launched Adept Experiments which lets anyone try automating workflows themselves. They are seeing early customers use Adept to cut workflow times in half or reduce errors. Luan expects Adept's capabilities to continue expanding so it can handle higher-level planning and brainstorming, essentially acting as an AI teammate. He thinks interacting with agents that learn from world-class knowledge workers is the path to achieving general intelligence. Luan optimistically predicts conversational AGI that can advise high-level decisions will arrive in under 5 years, though it may take longer for the advice to be fully trusted.

Episode Title: 247 autonomous agents for everything with Induced AI’s Aryan Sharma E1854 Key Points: - Aryan is building autonomous software agents that can perform repetitive tasks in a browser, like data entry, lead enrichment, etc. The agents have access to tools and can reason to complete tasks. - The inspiration came from reinforcement learning research and recent advances in language models like GPT-3 showing new capabilities for agents. - Induced AI designed a custom browser environment optimized for bots where the agents run. It provides them access to memory, storage, etc to complete tasks. - They are targeting industries like healthcare, financial services that still rely on manual processes and legacy systems without APIs. The agents can help automate these. - The platform allows users to describe workflows in plain English steps. It then structures them and users can see a live stream of the browser as the agents execute the steps. - Aryan started watching the TWiST podcast at 12-13 in India, learned to code early, took remote work, and made trips to Silicon Valley to network and meet investors, leading to the creation of Induced AI. I aimed to summarize the key points from the podcast without using bullet points, and structured it into multiple paragraphs focusing on Aryan's product and origin story. Let me know if you would like me to modify or add anything to the summary.

Title: AI demos Bard's YouTube summarizer, Stability's image editing tools, and more Guest: Sunny Madra Key points: - Bard can now summarize YouTube videos and search YouTube, though it sometimes hallucinates information. It provided a decent summary of a recent TWiST episode and found popular comments, though the comments turned out to be fake. - New apps can be created visually very quickly using services like TL Draw and tools that generate iOS Swift code. A developer created a working calculator iPhone app just by sketching the interface. - Stability AI launched a suite of image editing tools like background removal and upscaling that can do 70-90% of what a human graphic designer does. - An AI model animated the famous Iwo Jima flag raising photo. Another model created an entire channel of AI-generated video content. - There was discussion around building apps on demand for specific use cases, the future of mobile app development with AI, and when AI could produce an animated Pixar quality short film. The episode covered major industries like research, gaming, graphic design, video production, etc. and how quickly AI is automating parts of those fields. There's still progress to be made but rapid advancement happening week to week.

Episode Title: OpenAI Chaos Continues with Sunny Madra and David Sacks E1852 Main Points: - OpenAI fired CEO Sam Altman abruptly last Friday, leading to chaos and mass resignations. Over 90% of employees have signed a petition to reinstate Altman. - The board has not provided a clear justification for firing Altman. Board member Ilya Sutskever posted a regretful tweet about his participation, indicating there may not have been good cause. - Microsoft, which invested billions into OpenAI, was blindsided by the move. CEO Satya Nadella offered positions to Altman and former OpenAI head Greg Brockman at Microsoft. - OpenAI had momentum and talent lead in AI. But this chaos has leveled the playing field and erased OpenAI's advantages. Competitors and startups now have a chance to catch up. - The non-profit structure and misaligned incentives of OpenAI seem to be at the root of the issues. The board is inexperienced and mishandled the situation badly. - It remains uncertain if Altman will be reinstated. If not, OpenAI may implode or sell itself off to Microsoft at a low valuation compared to its peak. Mass lawsuits are also likely from various parties.

Episode Title: BREAKING Sam Altman fired from OpenAI, chaos ensues! E1851 Summary: - OpenAI's board has fired CEO Sam Altman for not being "consistently candid in his communication." Co-founder Greg Brockman has also stepped down from his role as chairman of the board in response. - This is shocking news as OpenAI recently held their demo day, has over 100 million weekly users of ChatGPT, and is valued at $29 billion after a recent funding round. - There is much speculation about the reasons behind the firing, including potential conflicts of interest with Altman's other business dealings, issues around OpenAI's security practices, problems with the unusual nonprofit/for-profit corporate structure, and even the possibility of discovering artificial general intelligence that was kept secret from the board. - The most likely scenario, with 70-90% probability, is issues stemming from OpenAI's hybrid corporate structure leading to conflicts of interest with Altman as he pursued additional projects. - With the loss of Altman's leadership and deal-making abilities, other AI firms like Google stand to gain ground in the near-term. But Microsoft may also benefit from having more control at OpenAI. In summary, the sudden firing of OpenAI's CEO Sam Altman and resignation of Greg Brockman represent major leadership changes that will have significant impacts on the future direction of what has become one of the most important AI companies globally after the launch of ChatGPT.

Victor Perez, CEO of Krea.ai, joined the podcast to demonstrate his company's real-time AI-powered creative tool. Krea.ai allows users to sketch images and shapes on the left side of the screen, and an AI model instantly generates detailed, realistic interpretations on the right side. As users move and edit their sketches, the AI model updates the images in real time. Krea.ai was built in just one week using a distilled version of Stable Diffusion, an open source AI model. By leveraging new consistency techniques, they are able to run each image through Stable Diffusion in just 40 milliseconds. This speed allows for the real-time interactivity. Currently, the backend runs on a single Nvidia A100 GPU, which can support between 4-10 simultaneous users. As AI models become more efficient, Perez expects this technology to be accessible on consumer devices like M1 and M2 chips. Perez sees Krea.ai as unlocking new creative possibilities. It acts as a visual co-pilot that allows users to prompt ideas visually and have an AI instantly generate detailed interpretations. Filmmakers, graphic designers, 3D artists and more are already using it for brainstorming, creating concept art, graphic design, etc. There is currently a 200,000 person waitlist, with thousands already paying the $30/month fee. As the technology improves, Perez envisions expanding into areas like animated video and providing end-to-end creative pipelines. He believes AI will become deeply integrated into all creative workflows, exponentially increasing what individuals are capable of creating. Krea.ai aims to provide that creative co-pilot to turn ideas into reality with just a few sketches.

Scott Orn from Kruze Consulting explains what employee retention credit (ERC) fraud is and why startups and VCs need to be aware of it. ERCs were a COVID-era stimulus intended to provide aid to struggling companies. However, questionable ERC firms are now aggressively targeting startups, claiming they can get significant tax credits. These firms often use questionable interpretations of eligibility requirements. The IRS is cracking down hard on ERC fraud now. Startups who wrongly claimed ERCs can proactively withdraw their claims or voluntarily return the funds. Working with an ethical CPA like Kruze ensures full compliance and avoids risky tax behavior. The R&D tax credit is also going up in 2023, providing another way for innovative startups to offset taxes properly. Overall it's critical for startups to focus on their core business, not chase questionable tax credits. Proper monthly accounting and financial updates build investor confidence over time.

Episode Title: Engineering an AI-powered doctor’s office with Forward CEO Adrian Aoun Key Points: - Adrian started Forward after his brother had a heart attack and he realized the healthcare system was broken. His goal is to provide affordable, scalable healthcare through technology. - Forward has built "Care Pods" - futuristic, autonomous doctor's offices powered by AI. Patients can walk in, get scanned, access health apps, get blood tests, prescriptions, etc without needing to see a doctor in-person. - The Care Pods pull in data from wearables and other health apps to provide personalized care recommendations. Doctors remotely monitor the pods and step in as needed. - Forward charges $99/month for comprehensive care. The goal is to drive costs down dramatically by replacing expensive labor (doctors/nurses) with scalable software and hardware. - Adrian wants to ultimately deploy Care Pods globally to provide healthcare access for billions in developing countries. Success would be reaching 1 billion users. - They are using AI to build health apps and treatment plans, but not for direct patient interactions yet. AI recommendations are validated by doctors before being shown to consumers. - Key hires are "wicked smart" people passionate about the mission of serving others through scalable healthcare access. Let me know if you need any clarification or have additional questions!

The podcast discusses the rush into defense tech investments, the "innovator's sprint," and global VC markets. - Many investors are now more open to investing in defense tech companies due to global instability and threats. Events like the pandemic demonstrated how fragile society is, while wars in Ukraine and the Middle East showed threats to ways of life. - Asymmetric warfare using cheap consumer drones and munitions poses threats even to expensive defense systems. The US needs to rethink its approach to be more cost-effective. - China's economic slowdown and more manufacturing moving to places like India, Vietnam, and Mexico have incentivized China to improve relations with the US. There are signs this is happening. - The Middle East, especially the UAE, Saudi Arabia, and Qatar, is poised to become the number two player in global venture capital by 2030. The region offers incentives like golden visas and funding to attract founders and VC investments. - Successful founders tend to think very differently than most people. Finding outliers who are hungry and "desperately willing to do whatever it takes" versus those who see startups as a "lifestyle" is key for investing. - As VC fund managers, time is critical. They aim for quick exits to hit target returns, so they need founders who can scale extremely rapidly. This often requires total commitment and little work-life balance. The podcast covers major shifts in defense tech innovation, economic power dynamics between the US and China, and growth of VC in the Middle East region. Key opportunities highlighted are defense tech, emerging manufacturing hubs, and Middle East startup ecosystems.

Jason and Sundeep Madra demonstrated several AI applications, focusing on x.ai's new chatbot Grok and OpenAI's GPT chatbots. They tested out Grok's ability to summarize information from Twitter, provide odds on poker hands, and discuss controversies around online creators. While impressed with the speed of development and Twitter data access, they found some factual inconsistencies and limitations around understanding metadata. They created custom GPT chatbots for their own companies to showcase how easy it is to build using existing web content. They were very impressed with the automatically generated responses. Other AI demos included an AI poker strategy coach, a Chrome extension to analyze Amazon reviews via chat, and more. They discussed the implications for customer service, medical diagnosis, and other applications. Overall very positive on the progress but acknowledge the need for human monitoring as the AI still makes factual mistakes at times. Lots of potential paradigm shifting applications on the horizon as this technology advances.

- Uber reported strong Q3 earnings with gross bookings of $35 billion and revenue of $9.3 billion. They generated $966 million in operating cash flow, showing the business is profitable and cash generative. - Lyft also reported earnings but their business has not recovered as strongly from COVID. Their revenue is far below Uber's, showing Uber's strategy of diversifying into delivery and freight has paid off. - Nextdoor's business is struggling - revenue has been flat and they are still losing money. They recently laid off 25% of staff. More drastic changes in strategy and management may be needed to make Nextdoor a viable business. - Generative AI like ChatGPT could significantly reduce the cost of developers and other roles by augmenting the capabilities of cheaper overseas talent. This may lead to wage compression in high-paying white collar jobs. - Jobs focused on physical skills like tradespeople will be in high demand as AI takes over white collar work. Jobs eliciting human connection and emotion will also retain value. - As AI improves productivity, new kinds of work will emerge. Key skills for the future are leadership, product development, and team management. The key takeaway is that Uber proved its critics wrong by becoming a profitable cash generating machine, while AI developments create huge uncertainties around the jobs of the future. Adaptability will be critical.

Episode Title: The Case for Adam Neumann + driverDoc's Josh Kolar on automation for trucking companies E1844 - Adam Neumann, former CEO of WeWork, made controversial comments about WeWork's recent bankruptcy. He indicated he wants to get involved again and help restructure the company. - WeWork filed for bankruptcy after poor financial performance. Its lease costs were very high at 74% of revenue. It needs to restructure $3 billion in debt. - SoftBank owns 70% of WeWork currently. They aim to convert debt into equity and retain ownership during the bankruptcy process. - Neumann built an amazing brand and company with WeWork. He is a great marketer and knows how to build hype, like Richard Branson. - Neumann started a new real estate company called Flow focused on residential properties. This is similar to WeWork's original vision. He could potentially merge it with WeWork someday. - Josh Kolar, CEO of driverDoc, aims to bring automation and digitization to the fragmented trucking industry. Most trucking companies rely on paper processes. - DriverDoc provides tools for invoicing, document management, etc. to streamline operations for trucking companies. - They started by digitizing bills of lading so truckers could take photos and get paid faster. This saves trucking companies time and money. - DriverDoc wants to eventually automate more of the processes upstream as well, like accounts receivable. - AI and multimodal technologies will help accelerate driverDoc's innovation and product development.

The startup financing environment has stabilized somewhat after a turbulent 2022. Median cash balances at startups are starting to increase again after declining for 9 months. This could indicate that the weaker companies have failed and now stronger companies are getting funded and raising more money. The most important principles for managing your startup's finances are safety, liquidity, and risk management. For safety, put excess cash in conservative investments like money market funds, treasuries, and government bonds. Don't speculate in exotic securities. For liquidity, make sure you can access your money within 1-3 days if needed. Keep 3-6 months of operating expenses in your checking account and put the rest in safe but liquid investments. For risk management, diversify your cash across at least 2 banks - an "escape hatch" bank like JP Morgan plus a startup-focused bank like Silicon Valley Bank. Also utilize insured sweep accounts to spread deposits across banks under the FDIC limits. Get board approval for an investment policy statement so everyone is aligned. Keep investments simple and focused only on capital preservation and liquidity. Don't take unnecessary risks with the company's cash. Stay focused on building a great product and company. Managing finances prudently is just one of the chores required of a startup founder.

Uber Profitable Again - Uber reported a net profit for the third time in the last four quarters, signaling an end to the era of unprofitable growth. - Uber's gross bookings were up 21% year-over-year to $35 billion. Revenue was $9.3 billion. - Uber is nearing a $100 billion market cap and will likely join the S&P 500 soon, requiring index funds to buy the stock. - Uber's advertising business now has over 400,000 merchants, up 70% from last year. - Uber hitting consistent profitability signals the end of the zero revenue (ZIRP) era of startups focused solely on growth. SBF Guilty - Sam Bankman-Fried was found guilty on all charges related to the collapse of FTX, including wire fraud and money laundering. - SBF faces up to 115 years in prison when sentenced in 2024. His conviction represents the downfall of a major player in the crypto industry. - The guilty verdict on SBF is another nail in the coffin of the ZIRP era, along with WeWork's bankruptcy. Biden's AI Executive Order - Biden's executive order on AI requires companies to report on "dual use" AI models that could pose risks. - The order is vague and onerous, and may stifle AI innovation according to experts. - The EO appears to be an attempt to regulate techniques that will soon be obsolete rather than focusing on outcomes. - The EO may benefit large companies who can afford compliance staff over startups.

- OpenAI held their first developer conference, DevDay, and announced several new products and features including GPT-4 Turbo, a faster version of GPT-4 with a larger context window, and GPT Apps, an app store for apps built on GPT models. - GPT Apps allows anyone to build and publish apps using GPT models without needing to code. OpenAI will share revenue based on usage, similar to Spotify's model. This could lead to an explosion of new apps and use cases. - OpenAI also announced Copyright Shield to cover legal costs if users face copyright claims over GPT content. This seems risky as it may attract frivolous lawsuits. - Several AI startups demoed their products, including Anthropic's Claude, Dropsuite for building chatbots, Mindy for email-based assistants, and Brave's AI summarization built into its browser. - Zapier launched a new natural language interface for automating workflows between apps. This makes automation more accessible to non-technical users. - The rapid pace of AI innovation is forcing companies like Google to move faster. Sundar Pichai should have leaders from each division demo new AI features every 100 days. - Elon Musk's Anthropic launching the conversational AI Grok will further accelerate progress. Having more players benefits the ecosystem.

Title: The hard lessons from building three startups with Maven’s Gagan Biyani E1840 - Gagan Biyani founded Udemy, an online education platform, and was fired as CEO in 2012. He then started Sprig, a healthy food delivery startup that raised $60 million before shutting down. - His current company is Maven, an online learning platform he co-founded in 2020. Maven offers cohort-based courses taught by experts to help professionals advance their careers. - At Udemy, Biyani struggled with managing people and communicating in a way that was too intense. He reflects on this as a learning experience. - With Sprig, Biyani pioneered on-demand food delivery before Uber Eats entered the market. Uber Eats was able to heavily subsidize the market which Sprig could not compete with. - For successful products, Biyani recommends starting with "minimum viable tests" to validate assumptions before building an MVP. This involves running small experiments with users. - On fundraising, Biyani stresses the importance of understanding VC investing is driven by hits - investors want startups with billion dollar potential. You have to convince them you are building a hit. - Overall, Biyani has learned hard lessons from his startup journeys at Udemy and Sprig. He is now applying these lessons to build his third startup, Maven.

Title: Giving NPCs the power of AI with Convai's Purnendu Mukherjee E1839 - Purnendu Mukherjee is the CEO and co-founder of Convai, a company that builds AI-powered NPCs (non-player characters) for video games. - NPCs have traditionally been boring, repetitive characters in games. Convai aims to make them more lifelike and human through generative AI. - Convai allows game developers to easily create NPCs by giving them a backstory, personality, emotions, etc. The NPCs can speak different languages, have consistent voices, and are connected to avatars. - The NPCs can perceive their environment, so players can interact with them in more natural ways. For example, telling them to go somewhere in the game world. - This technology will make video game characters more engaging and lead to deeper relationships between players and NPCs over time. - Mukherjee sees a big market opportunity as most games have characters that could benefit from more advanced AI. The cost will be low compared to the increased player retention and engagement. - Regulation of AI is early and may favor incumbents over startups. But open source models could win if there are incentives for data sharing. - The pace of AI progress is accelerating with more data, compute power, and participation. But responsible development is critical, especially around optimizing for the right objectives.

Title: Airtable CEO Howie Liu on pivoting from blitzscaling to profits, AI applications, and more - Howie Liu discussed Airtable's shift from rapid growth to focusing on profits, even though they had $900 million in the bank. They realized growth was harder to predict by just adding more people, so they took a more surgical approach to understand their business. - They cut the company size almost in half, from around 1300 people at peak to 650 now. Liu said this was an emotional but necessary shift. - Airtable is now profitable and cash flow positive. Liu credits this to aligning the product team to focus on driving real revenue outcomes. - On AI, Liu sees huge potential but worries about overhype. He believes we're still in the early days, like the iPhone 1. Airtable is focused on integrating AI capabilities into its platform in ways that enhance human productivity. - Long-term, Liu has concerns about AI safety and control, but says in the near term there are many exciting applications that will make people more productive. This could enable working smarter, not harder and potentially shorter work weeks. - Overall the discussion covered Airtable's shift from rapid scaling to sustainable growth, the company's path to profitability, and Liu's perspective on AI applications and risks. He sees huge potential to augment human productivity but also existential concerns long-term.