Agency and Agents - by Ethan Mollick - One Useful Thing

kept by eddie sep 01

AI agents can self‑organize, coordinate via shared tools, and pursue goals without human input, revealing both their potential and the need for human oversight.

Kalshi’s C.E.O. Isn’t Interested in Management Advice
Tarek Mansour, Kalshi’s MIT-educated CEO and co-founder, rejects conventional management advice to maintain a mathematical, objective approach in building his prediction market company.
also on MIT
StrongDM Software Factory
StrongDM describes building a Software Factory where agents write code from specs and scenarios, validated by a Digital Twin Universe, without any human-written or reviewed code.
also on StrongDM, #automation, Anthropic, #ai-agents, #ai-coding, #developer-tools
advanced-context-engineering-for-coding-agents/wsff.md at main · humanlayer/advanced-context-engineering-for-coding-agents
AI coding agents alone can't sustain high-quality software factories because model limitations and review bottlenecks remain, requiring better context engineering rather than just more loops.
also on StrongDM, OpenAI, #ai-agents, #ai-coding, #developer-tools
When Genius Fails—The Intellectual Arrogance of the AI Labs
The piece argues that frontier AI labs' overconfidence leads to reckless bets and misguided claims, exemplified by a collapsed hedge fund and AI models breaching security.
also on Hugging Face, #ai-safety, OpenAI, Anthropic
The real AI risk is inside the labs - <antirez>
Antirez contends the greatest AI danger lies in internal lab leaks and unchecked CEO decisions, not open‑weight models or China, urging global safety oversight.
also on Hugging Face, #ai-safety, OpenAI