OpenAI Town Hall with Sam Altman

Sam Altman answers questions and discusses the future of AI with builders from across the AI ecosystem. 00:00 Introduction 00:33 Will AI make software engineers obsolete or massively increase demand? (Jevons paradox) 02:18 If building is easy now, is go-to-market the real bottleneck? 04:35 What is OpenAI’s vision for agent builders and multi-agent UIs? 07:03 How can AI help close long-standing economic gaps like the gender wage gap? 09:53 Should models be specialized or general, given GPT-5’s tradeoffs? 12:27 How should developers think about “intelligence too cheap to meter” and model costs? 14:32 Why could custom interfaces accelerate micro-apps built just for individuals? 16:26 How do builders create durability when features can be replaced by model updates? 18:14 When will agents run long workflows autonomously without human intervention? 19:32 How can AI help improve the *quality of ideas*, not just execution? 23:06 Are we locking in foundations today that will be hard to change later? 25:08 Will AI eventually take over the full scientific research process? 29:50 How should we think about AI security and biocurity risks? 33:16 If AI makes individual learning instant, what happens to human collaboration? 38:18 What are the most underestimated failure modes as agents run production systems? 41:13 How should AI be introduced into early education and kindergarten? 43:00 Is there an inflection point coming for 3D reasoning and drug design? 43:47 Is university becoming limiting for AI builders, and do you still angel invest? 45:46 How will software engineering interviews change in an AI-first world? 51:40 How will human creative identity evolve alongside AI creation? 55:28 How will personalization and memory evolve across work and personal identities? 57:45 What is the most important skill to learn in the age of AI?