최신 글
4,338개정렬
27:16ENTudor Brindus on Finding and Reducing System JitterTudor Brindus · Jane StreetA memory ping-pong benchmark reveals how virtualization, interrupts, CPU frequency changes and speculation can disrupt low-latency systems.
15:20ENA No. 1 Gemini Ranking and the Limits of Early Model TestsA November 2024 look at preference rankings, reasoning tests and the future of scaling.
39:22ENA Fields Medalist Stuck for Two Hours. AlphaProof Climbed for Three Days.진행자 Sarah Guo, Elad GilThe DeepMind team behind AlphaProof tells Sarah Guo and Elad Gil how an AlphaZero-style agent, a Lean verifier, and test-time reinforcement learning reached IMO silver — and why the missing skill is still theory building.
1:27:25ENMarkus Villig on Building Bolt, Marketplace Growth and Self-Driving게스트 Markus Villig · 진행자 Harry StebbingsThe Bolt founder recounts an unusually lean start, difficult fundraising and global expansion, then explains his view of autonomous vehicles and the future of mobility.
11:10ENAI and Scientific Discovery: A 2024 Video’s Withdrawn StudyMatthew Berman · Forward FutureMatthew Berman discusses AI-assisted research and recursive improvement; the central preprint was later withdrawn over serious data-validity concerns.
46:35ENThey trained a real-time video model in 20 hours and skipped the game engineDecart CEO Dean Leitersdorf tells Sequoia how Oasis runs live on H100s, why ChatGPT is a fallen wall rather than a pain point, and why the next trillion-dollar company may be a generated experience.
1:00ENCan Rotating Three Tiles Reach Every Hexagon Tiling?A one-minute puzzle asks whether a simple local rotation can connect every rhombus tiling of a hexagon.
1:59:30ENHe Walked Away From Millions in OpenAI Equity — Then Wrote Four Rules for AGI Labs진행자 Nathan LabenzDaniel Kokotajlo and Dean W. Ball disagree on AI regulation, yet jointly argue that frontier labs need capability disclosures, public model specs, safety cases, and whistleblower protections.
1:42:56ENBuilding a Research and Writing Flow with CrewAIMatthew Berman, João Moura · Forward FutureMatthew Berman and CrewAI founder João Moura build and debug a two-crew workflow for planning and drafting educational articles.
1:04:54ENHe bet company equity that Transformer would last — then drew an AI map for 12,000 companies진행자 Jacob EffronDatabricks chief AI scientist Jonathan Frankle tells enterprises when to prompt, fine-tune, or pre-train, and why he still refuses to forecast the next architecture.
5:15:00ENDario Amodei Has Seen This Scaling Movie Enough Times게스트 Dario Amodei · 진행자 Lex FridmanAnthropic’s CEO puts numbers on frontier training costs and AGI, then hands the mic to the people who shape Claude’s character and look inside its weights.
1:45:59ENMatt Grimm on Defense Procurement, Anduril and the Cost of Autonomous Warfare게스트 Matt Grimm · 진행자 Harry StebbingsAnduril’s co-founder explains his case for procurement reform, the challenges of scaling hardware and his outlook on geopolitical risk.
36:49ENJensen Huang: NVIDIA Is No Longer Building Computers. It Is Building AI Factories진행자 Sarah Guo, Elad GilOn No Priors, NVIDIA’s CEO walks through hyper Moore’s law, NVLink, a 100,000-GPU cluster stood up in weeks, and why tokens are the new commodity.
1:00ENConnect Obsidian to a Local AI Model with LM StudioMatthew Berman · Forward FutureMatthew Berman outlines a way to connect an Obsidian note-taking plugin to a model running on your own computer.
0:35ENAn IBM Visit: Granite Models, InstructLab and Quantum ComputingMatthew Berman · Forward FutureMatthew Berman previews a longer discussion of IBM’s small AI models, model customization and quantum research.
0:28ENComparing Three Local Coding ModelsMatthew Berman · Forward FutureMatthew Berman introduces a sponsored comparison of DeepSeek Coder V2, Yi-Coder and Qwen Coder running on a workstation.
39:21ENSam Altman on Model Improvements, Agents and Startup Opportunities게스트 Sam Altman · 진행자 Harry StebbingsThe OpenAI CEO discusses where startups may build as models improve, and reflects on reasoning, infrastructure and leadership.
0:59ENCould an AI Arms Race Lead to Catastrophic Conflict?Two speakers compare AI risks with the nuclear age.
11:56ENIBM Granite 3.0 and InstructLab: Small Models for Enterprise AIMatthew Berman · Forward FutureMatthew Berman reviews IBM's 2024 model family and its approach to adding domain knowledge. IBM partnered on the video and invited him to its event.
1:15:17ENAn AI Arms Race Is Not Physics. It Is a Choice.진행자 Nathan LabenzNathan Labenz argues with Trump’s former NSC cyber director over China, chips, and whether words can stop a militarized AI race.
12:50ENComparing Three Local Coding Models on Games and Programming ChallengesMatthew Berman · Forward FutureMatthew Berman runs DeepSeek-Coder-V2 Lite, Yi-Coder 9B, and Qwen2.5-Coder 7B on a sponsored workstation. Their results vary by task and include manual corrections.
11:29ENChatGPT Search vs. Perplexity: A 2024 First LookMatthew Berman · Forward FutureMatthew Berman tests the newly launched search feature on podcast links, mistaken identities, restaurant maps, Factorio and stock questions.
55:01ENVlad Tenev on GameStop, Robinhood’s Growth and AI in Finance게스트 Vlad Tenev · 진행자 Harry StebbingsThe Robinhood CEO reflects on a turbulent trading crisis, company culture, financial products and the potential role of AI.
57:45ENThe last thing a totalitarian state can take is love — a reread of Orwell’s 1984진행자 Lex FridmanLex Fridman walks back through Oceania’s machinery of Newspeak, Doublethink, and Room 101, and argues that hate-fueled rebellion is not a way out.
더 보기4,338개 중 24개 표시









