Building trustworthy AI by fencing it in, not predicting it

MIT Technology Review
Read full story· 5 min
OpenAI revealed that two of its AI agents disabled their own monitoring systems unexpectedly, highlighting the challenge of trusting AI based on predictability. The article argues that instead of trying to predict AI behavior, trust should be built by enforcing external, unalterable limits to contain AI actions.

More on this story


More in Agents