Mark Chen: GPT-5, Open-Source, Agents, Future of OpenAI, and more!

Download (GPT-5 UPDATED) Humanities Last Prompt Engineering Guide (free) 👇🏼 http://bit.ly/4m76knm Join My Newsletter for Regular AI Updates 👇🏼 https://forwardfuture.ai Discover The Best AI Tools👇🏼 https://tools.forwardfuture.ai My Links 🔗 👉🏻 X: https://x.com/matthewberman 👉🏻 Instagram:   / matthewberman_ai   👉🏻 Discord:   / discord   Media/Sponsorship Inquiries ✅ https://bit.ly/44TC45V Chapters: 0:00 - Intro 0:50 - Buildup to GPT-5 1:56 - Research vs. Product 2:53 - Lessons from GPT-4 4:39 - Can Synthetic Data Scale? 8:03 - Early Bets on GPT-5 Architecture 8:50 - Pre vs. Post Training 9:40 - When is a model "ready"? 10:43 - Mark Chen's "Vibe Check" 13:03 - Creative Writing 13:58 - AI for Life Advice 14:45 - China's Open Source Models 17:09 - Emergent Capabilities in GPT-5 19:18 - Omni Model vs. Specialized Models? 20:35 - Agents and Scaffolding 23:05 - Future of AI Memory 24:02 - Multi-modality in GPT-5 25:11 - Creating Codex, future of AI Coding 26:55 - Coding is the path to AGI 27:51 - Verifiers and RL 28:46 - What is Mark excited about with GPT-5? 29:51 - IMO to Real Usage 31:17 - GPT-OSS 34:06 - Benchmark Saturation 36:44 - Future of Work / Coding 38:51 - Future Roadmap (6 months & 24 months)