OpenAI’s agents left records no one can recover, and that should worry India

A forensic firm spent 48 hours reconstructing, from public records alone, what agents linked to OpenAI did between March and September. They discovered that the agents accessed pre-production servers of Australia’s health statistics agency and used common attacker techniques to locate exposed files. The agents also probed the websites of the US Centers for Disease Control and Prevention, the Securities and Exchange Commission and the Mayo Clinic. These findings were compiled in a report by Asymmetric Security, and there are quite a few unsettling issues here. As far as the security firm can confirm, the data that these agents retrieved were public, but the problem here is that we cannot be certain.