Show Notes
OpenAI Research Scientist Noam Brown talks with AI Deep Dive host Rocket Drew about AI agents, reinforcement learning and what happens when increasingly capable agents begin reasoning, delegating and coordinating with each other. They dig into the infamous Hugging Face incident, the limits of AI “research taste,” OpenAI’s push toward AI systems that can help improve the next generation of AI, and the growing challenge of monitoring models’ chains of thought as they become more capable.
Related articles: Investigation into OpenAI Hugging Face Hack: https://www.theinformation.com/briefings/sen-josh-hawley-launches-investigation-openai-hugging-face-hack The Hugging Face Hack’s Chilling Postmortem: https://www.theinformation.com/newsletters/the-weekend/hugging-face-hacks-chilling-postmortem
Subscribe: YouTube: https://www.youtube.com/@theinformation
The Information: https://www.theinformation.com/subscribe_h
Sign up for the AI Agenda newsletter: https://www.theinformation.com/features/ai-agenda
Follow us: X: https://x.com/theinformation IG: https://www.instagram.com/theinformation/ TikTok: https://www.tiktok.com/@titv.theinformation LinkedIn: https://www.linkedin.com/company/theinformation/
Chapters: 00:00 - Introduction and Overview of AI Agents 04:08 - Reasoning and Reinforcement Learning 08:50 - Current Limitations, Astra, and Impact on Work 25:50 - Multi-Agent Systems and Delegation 31:04 - The Hugging Face Incident and Agent Coordination 36:05 - Game Theory and Agent Security 40:39 - AI Alignment, Monitoring, and Chain of Thought 50:07 - Future Outlook on AI Models and Interpretability
More from The Information's TITV
Is 1Password headed for an IPO? CEO David Faugno joins The Information's Akash to discuss the company's $500 million in annual recurring revenue, its new AI agent security tools and the growing cybersecurity risks posed by AI. Plus, Coco Labs co-founder and CEO Anusha Ramesh explains how her Jeff Dean-backed startup is tackling AI's design problem, and The Information's Laura Bratton breaks down why Salesforce customers could soon face new fees for using external AI agents like Claude. In The Editor's Cut, managing editor Laura Mandaro discusses our exclusive reporting on SB Energy's potential IPO delay, what Anthropic's anticipated public debut could mean for the market, and the mounting pressure on neocloud companies. Articles discussed in today's show: https://www.theinformation.com/newsletters/applied-ai/salesforce-customers-brace-new-fees-use-claude-apps
Mizuho Managing Director and Internet Equity Research Analyst Lloyd Walmsley talks with TITV Host Akash Pasricha about Meta’s slowing revenue growth, new AI monetization opportunities and the financing behind its enormous infrastructure spending. We also talk with The Information reporters Aaron Holmes and Leo Schwartz about an AI-discovered PHP security vulnerability, Microsoft CEO Satya Nadella’s pay cut and Andreessen Horowitz’s plans to drop “crypto” from its blockchain fund’s name. Morningstar Senior Equity Analyst Seth Goldstein joins us to discuss Tesla’s earnings and the prospects and risks of a SpaceX merger. Lastly, Legora Co-Founder and CEO Max Junestrand discusses the legal AI startup’s $550 million funding round, competition with Harvey and AI labs, and its path toward a potential IPO.
TMF Associates President Tim Farrar talks with TITV Host Akash Pasricha about Elon Musk’s proposed Tesla–SpaceX merger and the risks behind SpaceX’s fast-growing AI cloud business. We also talk with Wedbush Securities Senior Analyst Alicia Reese about Netflix’s slowing revenue growth and advertising challenges, and Different Co-Founder and CEO Ajeet Singh about his $48 million AI marketing startup and Anthropic’s reported compute commitments. Lastly, Eventual Founder Alex Keeney examines what prediction markets are signaling ahead of the midterms and how the elections could shape the industry.
Neuberger Senior Research Analyst Dan Flax talks with TITV Host Akash Pasricha about Meta’s AI strategy, Google’s cloud growth, Microsoft’s Azure business and why hyperscalers may keep spending aggressively despite higher interest rates. We also talk with Inspired Capital Founder and Managing Partner Alexa von Tobel about the firm’s new $450 million fund, Salesforce’s acquisition strategy and how AI agents could change marketing and commerce. Daytona Co-Founder and CEO Ivan Burazin explains the company’s $60 million raise at a $1 billion valuation and how its sandboxes help AI agents execute tasks, including work supporting Meta’s Muse model training. Lastly, The Information Finance Editor Meredith Mazzilli breaks down CoreWeave’s debt structure, the risks facing neocloud financing and the potential role of insurers in funding AI infrastructure.
Palo Alto Networks CIO Meerah Rajavel talks with TITV Host Akash Pasricha about securing AI agents and the risks of AI-generated code. We also talk with Modal Co-Founder and CEO Erik Bernhardsson about the company's $750 million funding round, $15.75 billion valuation and $1 billion annualized revenue run rate, and MoffettNathanson Managing Director Nick Del Deo about CoreWeave's debt, financing needs and power constraints. Lastly, Pentagon CTO Emil Michael joins The Information's Amir Efrati and Leo Schwartz to discuss the government's superintelligence task force, its dispute with Anthropic and the national security implications of the AI race with China.