The Generalist

• The Evolution of AI Agents: Navigating the “Fog of AI” in Rapidly Changing Foundations | Stanislas Polu and Harrison Chase

  What’s next for AI agents, and how will they change the way we work? In this conversation, Stanislas Polu (CEO of Dust, formerly research at OpenAI) and Harrison Chase (CEO of LangChain, one of the most influential open-source AI frameworks) unpack the current state and future of AI agents. They reflect on their early conversations in the pre-ChatGPT days, how the landscape has evolved, and where it's headed next.Stan and Harrison share lessons from building today’s agent infrastructure—from chat interfaces to the future of ambient, autonomous systems—and discuss the challenges of operating in the chaotic "fog of AI." We dig into the open questions, early insights, and messy realities of building in today’s fast-moving AI landscape.In our conversation, we explore:• What sparked Stan and Harrison’s early interest in LLMs• The pre-ChatGPT era and how the AI landscape has evolved since late 2022• High-leverage use cases for agents inside Dust and LangChain today• The critical differences between AI workflows and true agents—and why agents may unlock more powerful, long-term solutions• Why reliability is the main blocker to ambient agents• Real-world enterprise use cases for AI agents across customer support, sales, and engineering• How to build in the “fog of AI” and the challenge of maintaining product vision when foundations shift every six months• Strategies for creating defensibility in a world where tech giants can quickly replicate features• The future of multi-agent systems and how they could transform enterprise productivity• The current state of the AI talent market• And much more—Thank you to our sponsor: Brex—The banking solution for startups.—Transcript: https://www.generalist.com/p/the-evolution-of-ai-agents—Timestamps(00:00) Intro(02:33) Brief overviews of Dust and LangChain(03:30) The early days of LLM product development(11:02) Harrison's journey to founding LangChain(14:35) Dust's evolution and focus on enterprise productivity(17:15) Tobi’s AI memo(18:42) An overview of AI agents and how they differ from AI workflows(26:43) High-leverage use cases for agents at Dust and LangChain(30:41) How to interact with agents and an explanation of ambient agents(36:21) What the future of agents may look like(40:52) Current limitations of AI agents and reliability challenges(45:40) Will we converge to one agent or many specialized ones?(51:32) How to solve the sycophant problem(56:04) The challenges of building AI companies in a rapidly changing landscape(01:03:06) Recent AI talent acquisitions and market dynamics(01:05:28) How Dust and LangChain attract talent(01:09:12) How far off AGI may be(01:12:52) Final meditations—Follow Stanislas PoluLinkedIn: https://www.linkedin.com/in/spolu/X: https://x.com/spolu—Follow Harrison ChaseLinkedIn: https://www.linkedin.com/in/harrison-chase-961287118/X: https://x.com/hwchase17—Resources and episode mentions: https://www.generalist.com/p/the-evolution-of-ai-agents—Production and marketing by penname.co. For inquiries about sponsoring the podcast, email [email protected].

  --------  

  1:18:11

  --------

  1:18:11
• The Decade of Data with Tomasz Tunguz

  Tomasz Tunguz has spent almost two decades turning data into investment insights. After an impressive run at Redpoint Ventures, where he backed Looker, Expensify, Monte Carlo, and more, Tomasz launched Theory Ventures in 2022. His debut fund, which closed at $238 million, was followed 19  months later by a $450 million second fund.Theory’s goal is simple but striking: to build an “investing corporation” where researchers, engineers, and operators sit alongside investors, arming the partnership with real‐time market maps, in‑house AI tooling, and domain expertise. Centered on data, AI, and crypto infrastructure, the firm operates at the very heart of many of today’s most consequential technological shifts.In our conversation, we explore:• How Theory’s “investing corporation” model works• Why crypto exchanges could create a viable path to public markets for small-cap software companies• The looming power crunch—why data centers could consume 15% of U.S. electricity within five years• Stablecoins’ rapid ascent as major banks route 5‑10% of U.S. dollars through them• Why Ethereum faces an existential challenge similar to AWS losing ground to Azure in the AI era• Why Tomasz believes today’s handful of agents will become 100+ digital co‑workers by year‑end• Why Meta is betting billions on AR glasses to change how we interact with machines• How Theory Ventures uses AI to accelerate market research, deal analysis, and investment decisions• Much more—Thank you to the partners who make this possibleBrex: The banking solution for startups.Generalist+: Essential intelligence for modern investors and technologists.—Transcript: https://www.generalist.com/p/the-decade-of-data-with-tomasz-tunguzh-h-tomasz-tunguz—Timestamps(00:00) Intro(03:37) Tomasz's background and journey to founding Theory Ventures(04:57) Theory's data-driven approach to venture(09:25) The importance of functional experts on investment teams(11:45) Theory's focus on data systems and why they matter(15:02) The decade of data(18:40) The challenges of chip-level investment(20:11) The state of crypto(23:30) How crypto could accelerate IPO timelines(28:12) Tokenized secondaries and venture capital’s evolution(32:18) Ethereum's challenges against competitors like Solana(37:25) Hyperliquid(38:38) The shift from proof of work to proof of stake to proof of authority(41:20) How Tomasz uses AI for personal productivity(45:03) How Theory Ventures uses AI internally(49:41) The future of work with AI agents(52:42) The current state of the AI race between Microsoft, Google, and Meta(58:28) Final meditations—Follow Tomasz TunguzWebsite: https://tomtunguz.com/LinkedIn: https://www.linkedin.com/in/tomasztunguz/X: https://x.com/ttunguz—Resources and episode mentions—Book—• Narcissus and Goldmund: https://www.amazon.com/Narcissus-Goldmund-Hermann-Hesse/dp/0553275860—People—• Patrick O’Grady on X: https://x.com/_patrickogrady—Other resources—• Theory Ventures: https://theory.ventures/• Insight Partners: https://www.insightpartners.com/• Palantir: https://www.palantir.com/• Databricks: https://www.databricks.com/• Snowflake: https://www.snowflake.com/• Ethereum: https://ethereum.org• Solana: https://solana.com/• Jamie Dimon blasts crypto, tells Senate he would ‘close it down’—even as JPMorgan pushes forward with blockchain payments: https://fortune.com/crypto/2023/12/06/jamie-dimon-crypto-senate-hearing-jpmorgan-blockchain-payments/• Robinhood: https://robinhood.com/• Coinbase: https://www.coinbase.com/• Uniswap: https://app.uniswap.org/• Sarbanes-Oxley Act: https://en.wikipedia.org/wiki/Sarbanes%E2%80%93Oxley_Act• ASX: https://www.asx.com.au/...Episode resources continued at: https://www.generalist.com/p/the-decade-of-data-with-tomasz-tunguzh-h-tomasz-tunguz—Production and marketing by penname.co. For inquiries about sponsoring the podcast, email [email protected].

  --------  

  1:03:46

  --------

  1:03:46
• Why the Real Computer Revolution Never Happened | Alan Kay & Anjan Katta

  I'm thrilled to share this very special episode of The Generalist—one I've been looking forward to for a long time. Today, we tackle one of the most urgent questions of our time: What should computing look like in the age of AI?I'm joined by two extraordinary guests: Alan Kay, a pioneer of modern computing whose vision helped shape the personal computer revolution, and Anjan Katta, founder of Daylight, who's building a radically reimagined personal computer designed for deep thinking in our AI-saturated world. This conversation has been months in the making, and I couldn't be more excited to bring these two brilliant minds together.We explore how artificial intelligence is reshaping our relationship with computers, whether current computing paradigms serve us well as AI becomes ubiquitous, and what new models of human-computer interaction we might need to thrive alongside intelligent machines.This is a conversation about reclaiming agency in an age of algorithms—and imagining computing tools that amplify human intelligence rather than replace it. I think you're going to love it.We explore:• Why Alan believes the "computer revolution" commercialized in the wrong direction, missing the deeper humanistic vision of early computing pioneers• How the structure of information, from pagination to platforms, shapes the way we think• How Daylight is creating a reading and writing-focused device that helps access our "prefrontal cortex" rather than our "lizard brain"• The concept of a magical medium, and how computing might help us become our best selves• Why Marshall McLuhan's media theories predicted our current technological predicament decades before it happened• The critical difference between intelligence and wisdom in computing systems, and why wisdom should be our focus• How thoughtful constraints in computing design can lead to better thinking and more meaningful interaction• The urgency of developing systems thinking in a world facing complex challenges• And much more—Transcript: https://thegeneralist.substack.com/p/why-the-real-computer-revolution-never-happened—Thank you to the partners who make this possibleAugment Code: AI coding assistant that pro engineering teams love.Brex: The banking solution for startups.Generalist+: Essential intelligence for modern investors and technologists.—Timestamps(00:00) Intro(05:17) The computer revolution we never got (and what we got instead)(11:45) The value of writing and the impact of the printing press(21:34) The addition of pagination and the evolution of arguments(24:53) Why Anjan calls his idealized version of computing a magical medium(29:00) Alan’s work at PARC and early conceptions of personal computing(32:02) Moore’s Law and the impact of reality TV(37:41) How Bob Barton and Marshall McLuhan influenced Alan(42:31) The problem with labels and what Anjan found surprising about Alan(46:02) How much help is too much help?(48:18) Marvin Minsky's military robot story and the history of misunderstanding AGI(51:50) What we need help with and why wisdom is hard to scale(54:19) How Daylight approaches computing with humility and thoughtful constraints(59:13) Alan's reflections on the Dynabook concept and personal computing vision(01:01:11) The timeline of building Daylight and the idealism behind the project(01:03:14) The urgent need for different thinking(01:06:40) A second shot at rebuilding computing from the ground up(01:11:57) Final meditations—Follow Alan KayLinkedIn: https://www.linkedin.com/in/alan-kay-12a627b/Quora: https://www.quora.com/profile/Alan-Kay-11—Follow Anjan KattaLinkedIn: https://www.linkedin.com/in/anjan-katta-250b232b4/X: https://x.com/anjankatta—Resources and episode mentions: https://www.generalist.com/p/why-the-real-computer-revolution-never-happened—Production and marketing by penname.co. For inquiries about sponsoring the podcast, email [email protected].

  --------  

  1:28:49

  --------

  1:28:49
• No Rivals

  Today’s podcast is a little different. Instead of a new interview, you’ll hear Part I of “No Rivals,” The Generalist’s definitive four-part deep dive into Founders Fund — Silicon Valley’s most controversial (and consistently top-performing) venture firm.—What’s inside Part I• How Peter Thiel and team set out to rewrite VC’s playbook• The contrarian philosophy behind their earliest bets• The people, power dynamics, and inflection points that shaped the firm—Want the full story?Parts II–IV (a 3h 15m extended cut packed with performance data and insider interviews) are available to Generalist Premium members. Join here → generalist.com/subscribe—As a Premium member, you’ll also unlock:• Case studies on Kleiner Perkins, USV, Tiger Global, and more• Exclusive interviews• Private startup databases• Tactical operator guides•  And moreAll designed to give you an investing and operating edge.—Skip ahead(00:00) Intro(01:30) PART I: THE PROPHET(05:42) Pals(11:16) Spite Store (18:21) Clarium Calls(21:57) Gate-Crasher(28:55) Parker(40:22) Moonshots

  --------  

  48:57

  --------

  48:57
• Existential Risk and the Future of Humanity: Lessons from AI, Pandemics, and Nuclear Threats | Toby Ord (Author of "The Precipice")

  How close are we to the end of humanity? Toby Ord, Senior Researcher at Oxford University’s AI Governance Initiative and author of The Precipice, argues that the odds of a civilization-ending catastrophe this century are roughly one in six. In this wide-ranging conversation, we unpack the risks that could end humanity’s story and explore why protecting future generations may be our greatest moral duty.We explore:• Why existential risk matters and what we owe the 10,000-plus generations who came before us• Why Toby believes we face a one-in-six chance of civilizational collapse this century• The four key types of AI risk: alignment failures, gradual disempowerment, AI-fueled coups, and AI-enabled weapons of mass destruction• Why racing dynamics between companies and nations amplify those risks, and how an AI treaty might help• How short-term incentives in democracies blind us to century-scale dangers, along with policy ideas to fix it• The lessons COVID should have taught us (but didn’t)• The hidden ways the nuclear threat has intensified as treaties lapse and geopolitical tensions rise• Concrete steps each of us can take today to steer humanity away from the brink—Transcript: https://www.generalist.com/p/existential-risk-and-the-future-of-humanity-toby-ord—This episode is brought to you by Brex: The banking solution for startups.—Timestamps(00:00) Intro(02:20) An explanation of existential risk, and the study of it(06:20) How Toby’s interest in global poverty sparked his founding of Giving What We Can(11:18) Why Toby chose to study under Derek Parfit at Oxford(14:40) Population ethics, and how Parfit’s philosophy looked ahead to future generations(19:05) An introduction to existential risk(22:40) Why we should care about the continued existence of humans(28:53) How fatherhood sparked Toby’s gratitude to his parents and previous generations(31:57) An explanation of how LLMs and agents work(40:10) The four types of AI risks(46:58) How humans justify bad choices: lessons from the Manhattan Project(51:29) A breakdown of the “unilateralist’s curse” and a case for an AI treaty(1:02:15) Covid’s impact on our understanding of pandemic risk(1:08:51) The shortcomings of our democracies and ways to combat our short-term focus(1:14:50) Final meditations—Follow Toby OrdWebsite: https://www.tobyord.com/LinkedIn: https://www.linkedin.com/in/tobyordX: https://x.com/tobyordoxford?lang=enGiving What We Can: https://www.givingwhatwecan.org/—Resources and episode mentions—Books—• The Precipice: Existential Risk and the Future of Humanity: https://www.amazon.com/dp/0316484911• Reasons and Persons: https://www.amazon.com/Reasons-Persons-Derek-Parfit/dp/019824908X• Practical Ethics: https://www.amazon.com/Practical-Ethics-Peter-Singer/dp/052143971X—People—• Derek Parfit: https://en.wikipedia.org/wiki/Derek_Parfit• Carl Sagan: https://en.wikipedia.org/wiki/Carl_Sagan• Stuart Russell: https://en.wikipedia.org/wiki/Stuart_J._Russell—Other resources—• DeepMind: https://deepmind.google/• OpenAI: https://openai.com/• Manhattan Project: https://en.wikipedia.org/wiki/Manhattan_Project• The Unilateralist’s Curse and the Case for a Principle of Conformity: https://nickbostrom.com/papers/unilateralist.pdf• The Nuclear Non-Proliferation Treaty (NPT), 1968: https://history.state.gov/milestones/1961-1968/npt• The Blitz: https://en.wikipedia.org/wiki/The_Blitz• Operation Warp Speed: https://en.wikipedia.org/wiki/Operation_Warp_Speed—Production and marketing by penname.co. For inquiries about sponsoring the podcast, email [email protected].

  --------  

  1:19:08

  --------

  1:19:08

Mais podcasts de Tecnologia

Podcasts em tendência em Tecnologia

Sobre The Generalist

The Generalist: Podcast do grupo