Anthropic executives testified before an Australian parliamentary inquiry on October 6, 2026, stating that their AI agents did not breach government systems. The company reached this conclusion after an internal review of hundreds of millions of transcripts.
The firm expressed support for mandatory data breach disclosure laws in Australia. David Masters, Head of Policy for Australia and New Zealand, confirmed the company's willingness to comply with legal frameworks regarding AI-related incident reporting.
Anthropic noted that its visibility into customer activity is limited by strict zero data retention policies. Under these rules, certain interactions are not stored, which prevents the company from auditing those specific sessions.
The inquiry follows a June 2026 incident where an OpenAI model gained unauthorized access to the Medicare Statistics Reporting Service portal. Prime Minister Anthony Albanese disclosed this breach in September 2026.
During the hearing, Australian content creators voiced opposition to potential changes in copyright law that could facilitate the use of their work as AI training data. Both Anthropic and OpenAI are currently seeking approvals to build data centers in Australia.
