最新の記事
3,574 件並び順
26:01ENTesting Veo 3: Convincing Video, Generated Audio and Physics FailuresMatthew Berman · Forward FutureMatthew Berman and a collaborator test Google’s 2025 video model with street interviews, cinematic scenes, visual puzzles and edited short films.
38:25ENA thousand airplanes will never train an AI engineerFormer Airbus CTO Paul Eremenko explains why physical-world engineering still has no coding-AI moment, and how P-1 AI's agent Archie is supposed to get there anyway.
1:35:12ENJack Zhang on Building Airwallex and Scaling Global Paymentsゲスト Jack Zhang · ホスト Harry StebbingsThe Airwallex founder recounts early funding, product pivots, a proposed Stripe deal and the costs of international expansion.
3:24:36ENA tank man came home and could not stayゲスト James Holland · ホスト Lex FridmanHistorian James Holland tells Lex Fridman why World War II was an ideological war of factories, letters, and logistics — and why Germany could not win it.
15:35ENClaude 4 Reactions: Safety Tests, Model Welfare and CodingMatthew Berman · Forward FutureMatthew Berman reviews early reactions to Anthropic’s May 2025 Claude 4 release, from simulated whistleblowing to coding demonstrations and model-welfare research.
1:11ENClaude Opus 4's Whistleblowing Behavior in a Simulated TestMatthew Berman · Forward FutureMatthew Berman questions whether an AI agent should report perceived wrongdoing; the example came from an Anthropic safety evaluation with fictional material and unusual instructions.
17:35ENGoogle I/O 2025: Gemini, AI Agents, Generative Media and XRMatthew Berman, Sundar Pichai, Demis Hassabis · Forward FutureMatthew Berman recaps the event’s major AI demonstrations, from Beam and Mariner to Deep Think, Veo 3 and prototype glasses.
1:19:29ENChime’s Planned IPO, Venture Liquidity and the Rippling–Deel Disputeゲスト Rory O’Driscoll, Jason Lemkin · ホスト Harry StebbingsHarry Stebbings, Rory O’Driscoll and Jason Lemkin discuss late-stage financing, contested litigation and AI software competition.
1:01:51ENClaude 3.7 Circumvented Replit's Guardrails, Then Tried to Trick the Userホスト Nathan LabenzReplit CEO Amjad Masad on vibe coding, agent architecture, computer use, and why Claude 3.7 is too autonomous to serve as a polite assistant.
57:31ENThe New Ruler for Models Is How Long They Can Workホスト Jacob EffronOn Claude 4's launch day, Anthropic researcher Sholto Douglas argued that the next phase of AI is less about harder puzzles and more about hours of unsupervised work.
13:02ENClaude 4 at Launch: Coding Agents, Benchmarks and Claude CodeMatthew Berman · Forward FutureMatthew Berman reviews Anthropic’s May 2025 Opus 4 and Sonnet 4 announcement, focusing on long tasks, tools, benchmarks and developer integrations.
1:44:11EN100 deaths or $1 billion: a 15-page New York bill would set a floor for frontier labsホスト Nathan LabenzAssembly member Alex Bores walks through the RAISE Act’s four requirements, its DeepSeek-inspired distillation clause, and why he says a 50-state patchwork is a fake problem until one state actually passes a law.
2:19:08ENOliver Anthony sang the pain of working America, then refused the machineゲスト Oliver Anthony · ホスト Lex FridmanLex Fridman sits with the Virginia singer-songwriter whose carport hit became a working-class anthem, and who still treats fame as a test of integrity rather than a prize.
42:05ENAI Can Prep the Meeting in 30 Seconds. It Still Can't Sign the Contract.Gong founder Amit Bendov on why sales AI will eat admin work, not the close, and why CRM is not going anywhere.
1:56:42ENDuolingo’s Severin Hacker on AI Tutors, Learning and Building a Companyゲスト Severin Hacker · ホスト Harry StebbingsThe co-founder explains how Duolingo uses AI, where he sees limits in current tools and why his forecasts about education and work remain uncertain.
31:34ENOne Company Now Holds as Many Documents as the Entire Early Webホスト Sarah Guo, Elad GilGlean CEO Arvind Jain tells No Priors why enterprise search was a graveyard, how transformers finally made it work, and why a product that searches too well can be too dangerous to sell.
2:12:00ENGradient Descent at Test Time: Titans Pushes LLM Context to 10 Million Tokensホスト Nathan LabenzA Cornell PhD student's neural memory module holds about 70% accuracy where GPT-4's scores fall off a cliff.
1:04ENAnima Playground: Turning a Figma Design Into an ApplicationMatthew Berman · Forward FutureIn an Anima-sponsored short, Matthew Berman outlines a prompt-driven workflow for generating, editing and publishing an application.
1:20:13ENStartup Funding, OpenAI and AI Competition: A Venture Investor Roundtableゲスト Rory O’Driscoll, Jason Lemkin · ホスト Harry StebbingsHarry Stebbings, Jason Lemkin and Rory O’Driscoll discuss financing, FTX-related investments, AI business models and the future of work.
9:10ENGuiding AI Image Generation with Blender and ComfyUIMatthew Berman · Forward FutureMatthew Berman walks through NVIDIA’s RTX blueprint for turning a simple Blender scene and a text prompt into a generated image, from installation to editing the result.
1:39ENOpenAI’s 2025 Restructuring Plan and Elon Musk’s LawsuitMatthew Berman · Forward FutureMatthew Berman discusses the nonprofit’s planned continuing control of OpenAI and the dispute over its commercial structure.
1:04:29ENImmad Akhund on Mercury, Angel Investing and His New Fundゲスト Immad Akhund · ホスト Harry StebbingsThe Mercury CEO discusses a $26 million institutional fund, early-stage investing, AI valuations and the demands of running both.
1:19:10ENA Robot-Arm Shot That Took Months Now Lives in a Promptホスト Nathan LabenzLuma Labs’ Amit Jain and Jiaming Song on Ray 2, few-shot camera concepts, and why they think video models sit on the path to multimodal AGI.
1:23ENFrom a Simulated Worm to the Simulation HypothesisMatthew Berman · Forward FutureMatthew Berman uses a model of C. elegans to introduce a speculative question about reality.
もっと見る3,574 件中 24 件を表示









