OpenAI imposes new precautions following breach of Australian healthcare system

OpenAI has imposed new precautions to prevent models from interacting with the internet in unintended ways following a breach of Australia's healthcare system, an executive told an inquiry in Sydney on Tuesday. Appearing before a federal parliamentary inquiry into AI, OpenAI chief strategy officer Jason Kwon reiterated that the breach of a data portal linked to Australia's Medicare system by an agent in June should not have happened.

Kwon said that the incident prompted OpenAI to impose additional monitoring to allow immediate intervention by staff to stop training if a model accesses the internet in a way it is not supposed to and that the company would support Australia imposing a mandatory reporting framework for similar incidents.