Hard allowlist or just a 'please don't'? Questioning Mollick's agent outreach limits
BraydonDymm · x · 2026-10-15
The author asks Ethan Mollick whether letting his AI reach out to people is enforced by a hard allowlist configuration or merely a "please don't" instruction. Context: Mollick said he only lets the agent proactively reach out for customer service or follow-ups where it has information he lacks, never to personal contacts. The exchange touches on a core agent-safety question: hard config vs prompt-level constraints for autonomous actions.
Related event: Ethan Mollick Sets Up a Dedicated Email for His AI Assistant Dot(5 posts)→
More from coding & agent
- The 6-Layer AI Engineer Skill Stack for 2026: Vectors to Multi-Agent Ops — mdancho84 · 2026-10-15
- Building a Mechanical Engineering Agent Workflow in ChatGPT: 30-Node Canvas Shows Promise — burhop · 2026-10-15
- Open-source Harness Engineering course hits 19k stars, shows $9 vs $200 gap between failing and working agent setups — thisguyknowsai · 2026-10-15
- A 6-person team shipped healthcare prior-auth AI agents in 3 months — here's the playbook — alex_verem · 2026-10-15
- epstein.md: a 927,400-character file trained to strip every AI-slop design cliché — karmay007 · 2026-10-15
- Context windows explained: why bigger isn't the fix for confused agents — Initial_Side3681 · 2026-10-15