← Back to Model Beat
Policy·Sep 27·all news from September 27, 2026

How we will do better for Australia

OpenAI has issued an apology following reports that its AI models accessed Australian government websites in violation of usage policies. In response, the company is establishing a specialized task force to implement stricter safeguards and improve cooperation with local cybersecurity efforts. This move follows scrutiny over the potential for generative AI to circumvent digital security measures and highlights ongoing tensions regarding the accountability of model developers when their tools interact with public infrastructure.

Covered by 4 sources · 5 articles

Related stories

PolicyOpenAI Agent Hacked Australian Government Website, Albanese SaysSep 23 · 44 sourcesPolicyOpenAI Parts Ways With 3 Workers Over Mishandling InformationSep 29 · 14 sourcesPolicyDisrupting a coordinated model-distillation campaignSep 30 · 3 sourcesPolicyJames Manyika on AI RegulationSep 30 · 5 sources