No Priors

Coinbase co-founder and CEO Brian Armstrong joins Elad Gil to lay out Coinbase's vision for an 'Everything Exchange' encompassing tokenized stocks, stablecoin payments, and prediction markets. He argues AI agents will become the dominant economic actors on crypto rails, making agent-to-agent commerce a fundamental shift. Armstrong also unveils New Limit, his new longevity company pursuing epigenetic reprogramming to reverse age-related decline, and discusses early therapeutic programs and the path toward clinical trials.

Arm CEO Rene Haas joins Elad Gil and Sarah Guo to discuss Arm's expanding role in AI compute. He explains how the company is moving from IP licensing to manufacturing physical CPUs like the Arm AGI CPU for Meta, and why CPUs remain vital for AI workloads alongside GPUs. The conversation covers supply chain bottlenecks, the strategic importance of US semiconductor manufacturing, SoftBank's capital strategy, and future robotics opportunities for Arm.

Eon co-founders Ofir Ehrlich and Gonen Stein join Elad Gil to discuss the shifting value of enterprise data in the AI era. They argue that while compute and models get attention, proprietary data is a company's real protective moat. The conversation covers how Eon transforms cloud backup into an AI-ready data foundation by mapping, classifying, and providing access controls to scattered enterprise data. They explore the emerging security threat of rogue AI agents with legitimate credentials, and contrast the breakneck speed of AI adoption with the slower pace of the cloud transition.

Max Hodak, CEO of Science Corporation, joins Sarah Guo to discuss PRIMA, a retinal implant designed to restore functional vision for people who have lost their sight. They explore the convergence of biotech and computation, treating the brain as a system that can be repaired or augmented. The conversation extends to consciousness, identity, and whether human experience can persist as our biological hardware changes, as well as surprising parallels between AI models and biological brains.

Erik Allebest, co-founder and CEO of Chess.com, joins Sarah Guo to explain how a centuries-old game remains popular in an era of unlimited entertainment options. He recounts buying the Chess.com domain from a bankruptcy sale in 2005 and building a MySpace-like community that now has 10 million daily active users and 250 million registered members. The conversation covers the role of private equity investors General Atlantic and CVC in strengthening the platform, the cultural resurgence of chess, and the persistent challenge of cheating when machines can beat humans. Allebest also shares how Chess.com is using AI internally and in its product, along with founder advice and predictions about AGI and ASI.

Sarah and Elad explore the current AI landscape, discussing founder ambition in the face of AI lab dominance, the dynamics of building trillion-dollar companies, and shifting market sizes. They tackle exit frameworks, the looming promise of ASI and potential researcher burnout, compute bottlenecks, and how regulatory capture may drive ecosystems from California to Texas.

Melisa Tokmak, founder and CEO of Netic, joins Elad Gil to discuss how Netic’s autonomous AI platform acts as an intermediary between companies and customers, deploying agents to handle essential services such as emergency home repairs, hospitality, and pet care. Over 70% of Netic’s customers interact first with AI, a figure that reflects the system’s ability to absorb complex, real-world workloads that traditionally required large human support teams. Melisa also explains why she chose to build a scalable product company rather than an AI roll-up, why she believes robotics won’t catch up in these industries soon, and why large frontier AI labs are not competitive threats. The conversation covers private equity’s shift toward requiring measurable AI ROI and Melisa’s optimism about AI’s impact on education.

DoorDash co-founders Andy Fang and Stanley Tang discuss how autonomous tech and AI are reshaping delivery and consumer behavior. They cover the rollout of Ask DoorDash, a natural-language interface that increases grocery orders and restaurant discovery; Dot, their in-house autonomous delivery robot operating in Phoenix for two years; the 'first and last 100 feet' problem; and why multimodal strategies are critical. The founders explain why they believe more human Dashers will be needed as autonomy scales, not fewer.

Glenn Fogel, CEO of Booking Holdings, discusses scaling Priceline from the dot-com crash into a $100B+ global travel marketplace. He explains how AI agents like Penny transform travel planning and customer service, Booking's $700M+ annual AI investment strategy, the durability of their platform's scale, and why upskilling employees is critical as AI reshapes travel industry jobs.

Isaiah Taylor, founder and CEO of Valar Atomics, discusses how his company is shifting nuclear energy from theoretical designs to operational reality by building and iterating on advanced reactors. He explains why the US stopped building reactors in the 1970s, how Valar used a Department of Energy pathway (revived by executive order) to develop their advanced reactor, and why cheap, abundant atomic energy could fundamentally improve human quality of life. The episode includes a control room tour and discussion of powering an NVIDIA Blackwell chip directly from their live reactor.

OpenAI research scientist Noam Brown joins Sarah Guo to discuss his essay on why AI benchmarks are broken when models scale test-time compute. They explore how today's models can reason for weeks or months on complex tasks like math conjectures and poker solving, and why static benchmark grids fail to capture this capability. The conversation covers safety implications of variable compute budgets, the limits of recursive self-improvement, and the future of multi-agent collaboration.

At 66, former Cadence CEO Lip Bu Tan took over Intel to transform it from a legacy chip manufacturer into an agile, customer-focused company competing in agentic AI and inference. Tan discusses his strategy to strengthen Intel's balance sheet through strategic investments from Nvidia, SoftBank, and the US government, his product roadmap centered on CPUs for AI workloads, collaboration with Elon Musk on Terafab, and how he's applying startup culture principles—faster decisions, accountability, customer obsession—to reinvent a 50-year-old institution.

Biohub co-founders Mark Zuckerberg and Priscilla Chan, alongside Head of Science Alex Rives, discuss their $500 million virtual biology initiative integrating frontier AI with wet-lab work to build predictive models of cells and proteins. They announce ESMFold2, an open-source engine for digital protein and antibody design, and explain why Biohub operates as a nonprofit rather than a venture-backed startup. The conversation explores how hierarchical simulations could enable mechanistic, individualized medicine and accelerate the timeline for curing disease.

Satya Nadella discusses Microsoft's strategic positioning at the AI frontier, including multi-model harnesses, private evaluations as IP, and the reshaping of software engineering roles. The conversation covers how enterprises can operate effectively with frontier AI, the durability of SaaS models, data center ROI as community value, and AI's societal impacts including education and the token economy.

Sarah Guo and Elad Gil interview Maxim Bar Kogan, CEO of Onyx Securities, about securing autonomous AI agents at the enterprise level. Onyx builds a control plane that oversees agent permissions and contexts while managing latency, cost, and reliability tradeoffs. The conversation covers why current monitoring lacks sufficient context to detect agent intent, the limitations of proxy-based security, why Onyx trains its own models, and the critical need for vendor-independent AI governance as agents approach capability to manage critical infrastructure like power grids and water supplies.

Cerebras founder Andrew Feldman recounts the company's multi-year journey to pioneer wafer-scale AI computing, including surviving a chasm when demand hadn't yet caught up to its dinner-plate-sized chip. He explains how engineering breakthroughs deliver inference speeds 20x that of GPUs and shares how a $20 billion deal with OpenAI came together in just four weeks. Feldman also describes the mindset of a 'professional David' needed to challenge the chip industry's Goliaths.

Sarah Guo and Elad Gil interview Jacob Helberg, US Under Secretary of State for Economic Affairs, about Pax Silica, a 14-country economic security coalition designed to secure the entire AI supply chain from rare earth minerals to semiconductor manufacturing. Helberg discusses the creation of a forward-deployed industrial base in the Philippines, how Pax Silica differs from China's Belt and Road initiative, and the Trump administration's strategy for making AI policy durable across future presidencies.

Long Lake Management's acquisition of American Express Global Business Travel for $6.3 billion represents the first AI-driven take-private of a legacy platform. CEO Alexander Taubman explains Long Lake's strategy of acquiring and transforming businesses through their horizontal AI platform Nexus, rather than selling software to existing players. The conversation covers how AI automation drives both growth and employee satisfaction, the mechanics of AI-powered roll-ups, and Long Lake's vision of Amex GBT as a multi-decade compounding machine.

Tuhin Srivastava, CEO of Baseten, joins Sarah Guo and Elad Gil to discuss the explosive growth in AI inference demand and Baseten's role as the inference cloud. He explains why the application layer will persist as companies encode unique user data into custom models and post-train on proprietary signals. The conversation covers GPU supply constraints, Baseten's approach to multi-cloud orchestration across 18 clouds and 90 clusters, the shift toward longer GPU contracts, and the operational challenges of scaling an inference platform. Srivastava also shares his views on Chinese AI models, the dominance of custom inference workloads, and the future of multichip and concierge-like AI services.

Philipp Herzig, CTO of SAP, discusses how the 50-year-old enterprise software giant is navigating the AI transition for its 400,000+ customers. The conversation covers SAP's strategy for closing the AI adoption gap, the critical importance of data quality and verifiability in enterprise AI, the shift from predictive models to agentic systems, and how AI is forcing SAP to rethink its pricing model from per-user licenses to outcome-based approaches.

Roblox CEO Dave Baszucki explores the intersection of AI, physics simulation, and the future of human connection on virtual platforms.

Laura Deming, CEO of Until, dives deep into reversible cryopreservation — the technology that could enable organ banking and eventually full-body preservation.

NVIDIA CEO Jensen Huang reflects on the biggest surprises of 2025, including rapid improvements in reasoning and the profitability of inference tokens.

Sarah and Elad forecast 2026: foundation model competition, expected IPOs from OpenAI and Anthropic, and the emerging robotics wave.

Sarah and Elad review their favorite conversations and biggest takeaways from 2025, identifying the trends that will define 2026.

Harvey co-founder Gabe Pereyra discusses why legal AI is about more than individual productivity — it is about putting together complex client matters to make law firms more profitable.

ElevenLabs co-founder Mati Staniszewski on how AI voice technology is transforming how humans interact with technology and each other.

Cheng Chi and Tony Zhao discuss the gap between impressive robot demos and real-world deployment, and what it takes to bridge that gap.

Stanford professor and World Labs founder Fei-Fei Li discusses spatial intelligence — AI that understands and reasons about the 3D physical world.

CAIS executive director Dan Hendrycks discusses the real risks of advanced AI: not science fiction scenarios but concrete near-term dangers from misuse, misalignment, and concentration of power.

OpenAI's Brandon McKinzie and Eric Mitchell discuss the development of o3 and how reasoning models represent a paradigm shift in AI capability.

Brendan Foody of Mercor discusses how AI is transforming talent matching and hiring, creating a new kind of labor market infrastructure.

Dr. Shiv Rao of Abridge discusses how AI medical scribes are transforming clinical documentation, freeing doctors to focus on patient care.

Palo Alto Networks CEO Nikesh Arora discusses how AI is both the biggest security threat and the most powerful security tool.

Zach Dell discusses patterns in enterprise AI adoption, identifying which categories of AI enterprise software are finding product-market fit.

Flagship Pioneering founder Noubar Afeyan discusses how AI is transforming drug discovery and biotech, and why the most impactful applications are not the most obvious ones.

OpenAI's Isa Fulford discusses developer experience, API design, and how OpenAI is building the platform environment for millions of AI developers.

Sarah Guo and Elad Gil review the year in AI, covering the biggest breakthroughs, surprises, and disappointments. They discuss the gap between AI hype and AI deployment, the emergence of AI agents, and their predictions for 2025.

NVIDIA CEO Jensen Huang discusses the architecture of AI infrastructure, why data center design is the new semiconductor design, and why he bet NVIDIA's future on AI computing a decade before ChatGPT proved him right.

Andrej Karpathy discusses his vision for autonomous intelligence -- AI systems that can learn, reason, and act independently. He covers the limitations of current LLMs, the path from tool AI to autonomous AI, and why the next breakthrough will come from systems that can write and execute their own code.

Scale AI founder Alexandr Wang discusses why data quality is the limiting factor in AI development, how Scale AI became the data backbone of every major AI lab, and why the next frontier of AI requires not just more data but fundamentally better data.