Latest Interviews
Showing 1231–1245 of 1,332 transcripts.
Clear all filters- Lex Fridman1h 50m
David Patterson: Computer Architecture and Data Storage | Lex Fridman Podcast #104
David Patterson, Lex Fridman, John Hennessy
This comprehensive review traces the historical evolution of computing from the invention of microprocessors to the current slowdown of Moore's Law, highlighting key architectural shifts like the RISC vs. CISC debate and the rise of RISC-V as an open-source alternative. It details how industry giants and researchers have adapted to performance plateaus by developing domain-specific accelerators for machine learning and establishing transparent benchmarking standards through initiatives like MLPerf and RAID principles. The narrative concludes with David Patterson's insights on the synergy between research and teaching, alongside a realistic outlook that predicts quantum computing will not become commercially viable until around 2030 while immediate progress depends on software optimization and specialized hardware.
- Lex Fridman4h 9m
Ben Goertzel: Artificial General Intelligence | Lex Fridman Podcast #103
Ben Goertzel, founder of SingularityNet and architect of the OpenCog framework, synthesizes insights from literary figures like Stanislaw Lem and Philip K. Dick to argue for a decentralized AI future driven by biological and cognitive evolution. He details technical strategies for achieving Artificial General Intelligence through hybrid architectures while challenging corporate centralization via blockchain-enabled agent networks. Ultimately, Goertzel posits that humanity must embrace computational compassion and anti-death technologies to navigate the coming singularity, positioning AGI as the primary vehicle for solving existential threats like aging and resource scarcity.
- Lex Fridman2h 13m
Dawn Song: Adversarial Machine Learning and Computer Security | Lex Fridman Podcast #95
UC Berkeley professor Dawn Song outlines the persistent evolution of security threats, noting that while formal verification addresses system vulnerabilities, the human element remains the primary target for social engineering and adversarial machine learning attacks. In response, her research and startup Oasis Labs develop AI-driven defense agents, differential privacy mechanisms, and blockchain-based platforms to protect data ownership and ensure robust, privacy-preserving computing ecosystems. Ultimately, Song advocates for a transparent data economy where individuals control their information while leveraging program synthesis and international collaboration to advance artificial general intelligence.
- Lex Fridman1h 37m
Ilya Sutskever: Deep Learning | Lex Fridman Podcast #94
In this analysis of deep learning's evolution, Ilya Sutskever identifies the convergence of backpropagation innovations, GPU compute, and ImageNet data as the pivotal forces that unified the field by 2011. He explains how overparameterization drives the "double descent" phenomenon and argues that language understanding emerges from scaling architectures to model semantics rather than relying on innate grammatical priors. Looking toward artificial general intelligence, Sutskever advocates for a staged release strategy and proposes a governance model where AI systems internalize human values to act as benevolent agents aligned with collective flourishing.
- Lex Fridman1h 38m
Harry Cliff: Particle Physics and the Large Hadron Collider | Lex Fridman Podcast #92
Particle physicist Harry Cliff examines the mechanics of the Large Hadron Collider and the Standard Model during a 2019 Royal Institution talk, detailing how 27-kilometer accelerators probe quantum fields to validate the Higgs mechanism and search for supersymmetry. The discussion highlights the LHCb experiment's investigation of bottom quarks for potential new physics anomalies while outlining future engineering ambitions like the High-Luminosity upgrade and a proposed 100-kilometer Future Circular Collider. Furthermore, the dialogue underscores the critical role of machine learning in managing massive data streams and the collaborative international culture that drives these high-energy physics discoveries.
- Lex Fridman2h 9m
Dmitry Korkin: Computational Biology of Coronavirus | Lex Fridman Podcast #90
Professor Dimitri Korkin of Worcester Polytechnic Institute led a rapid, open-source initiative to reconstruct SARS-CoV-2 protein structures, leveraging computational genomics to identify conserved drug-binding sites and accelerate repurposing efforts. His research distinguishes viral "intelligent" efficiency from traditional biology, emphasizing how asymptomatic spread and minor genomic mutations drive transmission while shaping trade-offs between pathogenicity and contagion. This work underscores the critical role of global data sharing and agent-based modeling in refining containment strategies and advancing the development of universal vaccines against evolving strains.
- Lex Fridman2h 47m
Eric Weinstein: Geometric Unity and the Call for New Ideas & Institutions | Lex Fridman Podcast #88
Eric Weinstein argues that the COVID-19 pandemic has ended a 75-year period of global stability he calls the "Great Nap," exposing the fragility of electronic systems and the dangers of an establishment he describes as a "Distributed Idea Suppression Complex." He contrasts this institutional paralysis with his "Theory of Geometric Unity," a 14-dimensional mathematical framework he released to bypass academic gatekeeping that he claims stifles high-risk innovation. Weinstein advocates for a "neutron bomb" approach to these institutions, urging a new generation of outsiders to replace the current leadership to restore productive systems and enable progress in fields ranging from physics to interplanetary expansion.
- Lex Fridman1h 48m
David Silver: AlphaGo, AlphaZero, and Deep Reinforcement Learning | Lex Fridman Podcast #86
DeepMind researcher David Silver chronicles the evolution of reinforcement learning from early, rule-bound Go programs to AlphaGo Zero and MuZero, systems that achieve superhuman performance by learning through self-play without human data. This trajectory, highlighted by AlphaGo's historic 2016 victory over Lee Sedol and generalized across Chess and Shogi via AlphaZero, demonstrates that machines can master complex environments and exhibit creativity previously considered uniquely human. Silver argues that this progression formalizes intelligence as goal-oriented interaction, proving that with sufficient computational scale, artificial systems can surpass biological limitations in optimization and problem-solving.
- Lex Fridman1h 57m
Nick Bostrom: Simulation and Superintelligence | Lex Fridman Podcast #83
Nick Bostrom outlines the simulation argument, a probabilistic disjunction asserting that we likely inhabit a digital reality created by a post-human civilization, contingent on the extinction or lack of interest of advanced societies. The discussion extends to the implications of superintelligence and computationalism, positing that conscious simulation is feasible once civilizations achieve technological maturity through molecular nanotechnology and rapid AI self-improvement. Consequently, the speaker emphasizes the necessity of proactive AI alignment and risk management to navigate existential threats while exploring how a post-scarcity digital existence could redefine human values and meaning.
- Lex Fridman1h 39m
Anca Dragan: Human-Robot Interaction and Reward Engineering | Lex Fridman Podcast #81
Berkeley professor Anka Jagan explores the complexities of human-robot interaction by challenging standard prediction models with a bounded rationality framework that infers human intent through intuitive physics and active information gathering. Her research applies these principles to autonomous driving, where robots must negotiate ambiguous human behaviors and adapt to "leaked" environmental reward signals rather than relying on fixed objectives. Jagan advocates for treating robots as collaborative partners in a dynamic, game-theoretic system that continuously learns from human corrections and physical interactions to navigate unpredictable edge cases safely.
- Lex Fridman1h 35m
Vitalik Buterin: Ethereum, Cryptocurrency, and the Future of Money | Lex Fridman Podcast #80
Vitalik Buterin explores the strategic necessity of anonymity in Satoshi Nakamoto's creation of Bitcoin while contrasting it with the personal burdens faced by known figures like himself in the Ethereum ecosystem. The discussion details how quadratic funding and cryptographic consensus mechanisms like Proof of Stake address the economic challenges of public goods and decentralized coordination, evolving from early Proof of Work models to scalable sharded networks. Ultimately, the analysis positions these technologies not merely as financial alternatives but as essential backstops for individual sovereignty and global public goods provision in an era dominated by fiat currencies and state regulation.
- Y Combinator1h 50m
YC SUS: Eric Migicovsky hosts founder office hours
Eric Migicovsky, Eric Michakowski, JC, Mary, John, Jake, Nate, Chefy, Wee Susu, Armin, Amy, Lindy
Startup School facilitator Eric Michakowski leads concentrated office hour sessions where founders receive targeted strategic guidance from YC partners to solve specific scaling, product, and go-to-market challenges. Participants from diverse sectors including hardware, fintech, and biotech present their current bottlenecks to receive actionable advice on metrics, user acquisition, and regulatory compliance. The sessions are publicly recorded to foster peer support and provide a resource library of problem-solving strategies for the broader founder community.
- Lex Fridman1h 40m
Marcus Hutter: Universal Artificial Intelligence, AIXI, and AGI | Lex Fridman Podcast #75
Marcus Hutter, Lex Fridman, Jürgen Schmidhuber, Shane Legg
Theoretical computer scientist Marcus Hutter presents AIXI, a mathematically defined artificial general intelligence framework that combines Solomonov induction with sequential decision theory to establish a theoretical upper bound for optimal agency. To accelerate progress toward this goal, Hutter has increased the Hutter Prize for lossless compression to 500,000 euros, positing that intelligence is fundamentally equivalent to finding the shortest program capable of predicting and compressing data. This approach prioritizes formal definitions over heuristic methods to solve the reward problem through intrinsic curiosity and aims to enable autonomous systems to eventually resolve complex universal questions like the theory of everything.
- Lex Fridman1h 46m
Michael I. Jordan: Machine Learning, Recommender Systems, and Future of AI | Lex Fridman Podcast #74
Michael I. Jordan, Lex Fridman, Andrew Ng, Zoubin Ghahramani, Ben Taskar, Yoshua Bengio, Yann LeCun
Michael I. Jordan reframes the current state of artificial intelligence not as the engineering of human-like cognition, but as a nascent discipline focused on building large-scale decision systems, while explicitly rejecting premature claims of deep neurological understanding or full brain-computer integration. He distinguishes his approach from pure prediction by prioritizing decision-making under uncertainty and advocates for a shift from ad-based surveillance economies to direct producer-consumer markets that utilize game theory to align incentives with societal health. Jordan concludes that advancing this field requires a blend of rigorous mathematical frameworks, such as empirical Bayesian methods, and broad humanistic education to cultivate the empathy and collaboration necessary for solving unsolved challenges like natural language understanding.
- Lex Fridman1h 34m
Scott Aaronson: Quantum Computing | Lex Fridman Podcast #72
Scott Aaronson advocates reframing unanswerable philosophical questions into testable scientific inquiries, such as predicting human behavior within physical constraints, while explaining how quantum computers utilize superposition and interference to solve problems intractable for classical systems. He details the current transition through the noisy intermediate-scale quantum era, highlighting Google's 2019 supremacy demonstration and the critical engineering hurdles of decoherence and error correction required before practical applications like drug discovery or cryptographic threats become viable. Ultimately, Aaronson warns against hype surrounding quantum machine learning, urging a focus on verified quantum speedups and the substantial resources needed to move from theoretical models to reliable, error-corrected hardware.