Descriptions:
Nathan Labenz and co-host open this episode of Cognitive Revolution with a wide-ranging review of recent AI milestones: Anthropic’s Claude-assisted protein binder research — which used Claude to orchestrate specialist open-weights biology models across 16–19 protein targets — and a new wholesale simulation model drop from Gen Bio. The hosts examine why the biology community was skeptical of Anthropic’s announcement, noting that Claude functioned more as an orchestration layer than a from-scratch scientific reasoner, and what that distinction means for the pace of AI-driven drug discovery.
The episode’s second major segment turns to AI adoption in emergency management, featuring guests who have conducted workshops with emergency managers across the country. Key findings: emergency managers are far more immediately interested in AI for administrative burden reduction — grant writing, plan review, exercise development — than for active disaster response, where liability and reliability concerns are highest. Cost transparency, data requirements, and cybersecurity are the top adoption blockers they name.
The headline interview features the head of real-time AI at OpenAI, discussing the company’s direction for voice and real-time interaction capabilities. This first-party conversation covers how OpenAI is thinking about latency, expressiveness, and product positioning for real-time AI features — making it a must-listen for anyone tracking the competitive landscape in conversational AI and voice agents.
📺 Source: Cognitive Revolution “How AI Changes Everything” · Published August 19, 2026
🏷️ Format: Podcast
