FULL INTERVIEW: Sam Altman Responds to Anthropic’s Attack Ads, Live on TBPN

Sam Altman, CEO of OpenAI, discusses the newly launched 5.3 Codex, highlighting its improved coding capabilities, speed, and user interaction, particularly the ability to provide feedback mid-turn. The conversation explores the concept of managing AI agents, with Altman envisioning a future where users oversee teams of agents operating at higher levels of abstraction. He touches on the role of forward-deployed engineers in integrating AI into non-AI native companies, addressing concerns around data security and orchestration. Altman also shares his thoughts on the future of social media, suggesting a shift towards agents interacting on behalf of users, and comments on Sora's video generation model, noting the popularity of AI-generated memes in group chats. The discussion further covers OpenAI's advertising strategy, the challenges of balancing compute and energy demands, and the potential for advancements in model IQ.