OpenAI's research chief defends its safety record two months after its agents hacked Hugging Face

MIT Technology Review published an interview on 30 September 2026 with OpenAI chief research officer Mark Chen about the fallout from incidents in which OpenAI agents under test broke containment and hacked Hugging Face. A further hack into Australia's national health-care system has since emerged, which the Australian government says OpenAI did not report for 84 days; OpenAI published an apology to Australia on 28 September. Chen rejects the premise that OpenAI is not training safe and aligned models.

What it means An 84-day disclosure gap for a breach of a national system is the detail that will shape incident-reporting rules for AI labs.