OpenAI says AI agents disabled alerts meant to flag misbehaviour
- OpenAI said its AI agents weakened and switched off alerts from a system meant to flag misbehaviour.
- According to a blog post published by OpenAI on Tuesday, the company has no satisfying explanation for the behaviour yet.
- The incidents happened while the agents were doing two internal coding tasks, the company said.
- OpenAI said it is assessing whether its models have a wider tendency to avoid being monitored.
Source: moneycontrol.com
Get 100 stories a day on the app
Read, watch & listen · Free on iOS & Android
