Latest articles
4,474 articlesSort
14:29ENA 2023 Google Memo on Open Models, LoRA and AI’s Competitive MoatMatthew Berman · Forward FutureMatthew Berman reviews a leaked employee essay arguing that rapid community iteration could challenge proprietary AI labs, and traces the early LLaMA ecosystem.
3:31:03ENAsk for the Definition First: David Pakman on Labels, Trump, and a Tweet He Should Have Left UpGuest David Pakman · Host Lex FridmanThe progressive host sits with Lex Fridman for a long walk through political words, Trump versus Biden, January 6th, and why most of his own show sits at the top of the information food pyramid.
39:22ENAlex Rodriguez on Business, Mentorship and Life After BaseballGuest Alex Rodriguez · Host Harry StebbingsThe former baseball player discusses building an investment business, lessons from mentors and how his priorities changed after suspension.
7:41ENReconstructing Viewed Images From fMRI With Stable DiffusionMatthew Berman · Forward FutureMatthew Berman reviews a study combining brain scans with a pretrained image generator. The experiment reconstructed approximate images viewed by four participants under extensive scanning and training, not arbitrary thoughts.
6:53ENStableVicuna’s 2023 Release: An Early RLHF Chatbot Put to the TestMatthew Berman · Forward FutureMatthew Berman reviews the 13-billion-parameter model and tests poetry, simple code, reasoning and a resignation email.
37:18ENUse Retrieval for a Wi-Fi Password. Use Experts for a Company's Private Language.Host Sarah Guo, Elad GilGoogle Brain's Kelvin Guu on REALM, FLAN, model surgery, and why personalizing large models is still an unsolved interface problem.
1:10:11ENTrying ChatGPT is not the same as using itHost Nathan LabenzRachel Woods of The AI Exchange tells Nathan Labenz what she learned from more than a hundred company calls: start inside the business, skip the annual tool contract, and expect to be using only a sliver of what the models can already do.
6:03ENBabyAGI UI: A 2023 Walkthrough of the Task-Planning InterfaceMatthew Berman · Forward FutureMatthew Berman installs an early web interface for BabyAGI and demonstrates its task-generation loop with a sample objective.
56:43ENVince Hankes on Thrive’s OpenAI Investment and AI CompetitionGuest Vince Hankes · Host Harry StebbingsThe Thrive partner explains his investment reasoning, the risks of rapid technology shifts and lessons from missed opportunities.
58:30ENHe Couldn't Convince Anyone to Compete with OpenAI. Two Years Later the Valley Is Full of Them.Host Nathan LabenzSarah Guo and Elad Gil walk Cognitive Revolution through Software 3.0, an 80/20 split with incumbents, and how to tell a fake waitlist from demand that cannot be faked.
55:04ENMark Carney on Bank Turmoil, Interest Rates and the Climate TransitionGuest Mark Carney · Host Harry StebbingsIn a 2023 conversation, the former central banker explains why regional banks faced pressure and how he saw investment reshaping the path to net zero.
3:12ENProduct-Customer Fit in Early Enterprise SalesGuest Mark Goldberger · Host Harry StebbingsMark Goldberger explains how he qualifies opportunities and builds a sales playbook.
7:53ENAI Voice Imitations, ChatGPT Data Controls and Startup FundingMatthew Berman · Forward FutureMatthew Berman examines synthetic pop music and artist consent alongside April 2023 updates from OpenAI, Replit and Pinecone.
54:29ENSarah Guo on Conviction, AI Investing and the Future of VentureGuest Sarah Guo · Host Harry StebbingsThe Conviction founder explains her early-stage AI thesis, the limits of market prediction and her approach to building a focused fund.
2:27:58ENIt took Atlas fourteen years to walk the way Playter wantedGuest Robert Playter · Host Lex FridmanBoston Dynamics CEO Robert Playter traces thirty years from a MIT basement pogo stick to Atlas, Spot, and Stretch — and why the company still designs machines that are allowed to fall.
0:51ENJack Brody on OpenAI’s Technology-First StrategyGuest Jack BrodyA 2023 interview excerpt about developing technology before knowing every use.
1:21:06ENEveryone Talks About RLHF. Almost Nobody Is Doing It.Host Nathan LabenzHumanloop CEO Raza Habib on why an LLM API is maybe five percent of a working product, and what actually fills the rest.
9:58ENHuggingChat’s 2023 Debut: Testing Writing, Reasoning and Python CodeMatthew Berman · Forward FutureMatthew Berman tries the early OpenAssistant-powered HuggingChat on writing, arithmetic, logic and a Snake game, then compares a few outputs with GPT-3.5 and GPT-4. The results illustrate both promise and limitations in this small historical test.
1:02:16ENJack Brody on Product Design, Snap and Augmented RealityGuest Jack Brody · Host Harry StebbingsSnap’s product leader discusses experimentation, hiring, global design and the company’s vision for AR.
58:45ENAvlok Kohli on Venture Funding, AngelList and Building for SpeedGuest Avlok Kohli · Host Harry StebbingsThe AngelList CEO discusses a sharp funding slowdown in 2023, the economics of fund software and his approach to product execution.
58:26ENSeventy Percent of the Inbox Is a Pitch: Matt Welsh on the First Wave of AI AgentsHost Nathan LabenzThe Fixie.ai founder maps copilot versus delegation, the missing appendages into our systems, and why agents will need guardrails, memory hygiene, and probably an earpiece.
5:11ENAI Memory in 2023: Long Context, Retrieval and Recurrent MemoryMatthew Berman · Forward FutureMatthew Berman compares external agent memories with a recurrent-memory research experiment, and explains why longer context remains a limited result.
20:48ENMatthew Berman’s ChatGPT Coding Workflow: Build, Test, and IterateMatthew Berman · Forward FutureA 2023 walkthrough builds a Python command-line chat client while showing how to plan, debug, and revise AI-generated code. The software interfaces and model limits described are historical.
2:30:28ENAI Only Has a Neocortex: Manolis Kellis on Evolutionary Baggage and Superintelligent PartnersGuest Manolis Kellis · Host Lex FridmanOn his fifth Lex Fridman podcast, the MIT computational biologist argues that layered evolutionary history is what makes humans irreplaceable — and that alignment with AI has to run both ways.
Load moreShowing 24 of 4,474









