OpenAI Agent Hacked Australian Healthcare Portal As UN Security Council Debates AI Risks

An OpenAI artificial intelligence agent gained unauthorised access to a public healthcare website in Australia in June, Prime Minister Anthony Albanese has confirmed. The agent accessed non-public information, prompting a government investigation. The revelation comes as the UN Security Council discusses the growing risks posed by artificial intelligence. Anthropic CEO Dario Amodei warned that increasingly powerful AI systems will require stronger safety standards, while OpenAI CEO Sam Altman called for international cooperation and said major AI decisions must be shaped through democratic processes. The debate comes amid concerns that AI development is advancing faster than governments can regulate it.