Can AI Systems Police Themselves? The High-Stakes Gamble of AI Oversight
AI is making mistakes faster than humans can catch, and humans are using LLMs against each other to try and identify errors. Reworked says, relying on that alone is dangerous, and that while machines might track responsibility, they can't own it.
Senior Director, Leadership Ethics, Ann Skeet says, “Ultimately, some one person and the company that person works for will bear responsibility for the failures of AI systems."
The stakes extend beyond legal exposure, Skeet warned. "The legal battles that will be launched, if we settle for having AIs overseeing AI, will be legion, but the ethical considerations are even greater."
Ann Skeet, senior director, leadership ethics, quoted by Reworked.
Nov 10, 2025