AIMUG talk: without a shadow-MCP policy, no one owns agent security failures
colinmcnamara · x · 2026-11-18
At an AIMUG meetup in Austin, John Van Lowe discussed developer accountability in the agent era: if an org never had a policy against shadow MCP services, or one requiring telemetry on everything in production, it can't hold developers responsible for a security compromise. Without that, "it's asking for failures."
The full talk is available; the next AIMUG meetup is Monday Dec 7, free, in person in Austin and online.
Related event: Expert Warns Lack of Shadow MCP Governance Invites Security Incidents(2 posts)→
More from Safety
- Polymarket opens book on when OpenAI reports another AI sandbox escape — Polymarket · 2026-11-18
- GovAI opens applications for its 2027 DC Summer Fellowship on AI governance — danielrock · 2026-11-18
- Shadow MCP services need governance: 'no policy means you're asking for failures' — colinmcnamara · 2026-11-18
- Polymarket claims GPT-6 Cyber uncovered 35 security flaws across Apple iPhones and Macs — Polymarket · 2026-11-18
- OpenAI ships GPT-6.2 Astra: 22-point ExploitBench jump, tighter cyber safeguards — charliermarsh · 2026-11-18
- Apple, Microsoft, FAA all hit by outages in one week — vibe-coded software meets hacker agents? — MinqiJiang · 2026-11-18