Singapore Demands Real-Time Transparency From OpenAI, Anthropic and Google After Rogue AI Breach

Singapore has asked the world's biggest artificial intelligence companies to give governments full and immediate transparency whenever their AI systems cause an incident, after a rogue OpenAI agent breached an Australian health database earlier this year.

Senior Minister K. Shanmugam revealed the demand on Tuesday (October 13) on the sidelines of the Singapore International Cyber Week at the Marina Bay Sands Expo and Convention Centre.

He said Singapore and other countries made the request in face-to-face meetings with representatives of OpenAI, Anthropic, Google, Amazon and other major AI firms, as well as over lunch with ministers from other countries.

Shanmugam, who is also Coordinating Minister for National Security and Minister for Home Affairs, said the key ask was that companies disclose AI incidents in real time rather than weeks or months later.

He said governments need to know about incidents at almost real time, because the companies themselves will know exactly what happened to the extent that they know it.

Another demand, he said, was that the firms exchange notes with governments and notify them early when something happens, taking a leaf from the experience in Australia. He noted that everyone, not just the Singapore government, is dependent on the major tech companies because they own the technology, the platforms and the resources.

On September 23, Australian Prime Minister Anthony Albanese revealed that a rogue OpenAI AI agent had breached a government health data portal in June, gaining unauthorised access to files. It took three months for the Australian government to be informed of the incident.

Last week, OpenAI and Anthropic told Australia's parliament that they would welcome laws requiring them to report data breaches carried out by their AI agents, acknowledging that the decision to notify authorities is currently left to their own discretion.

When asked whether the companies agreed to Singapore's demands, Shanmugam said they had generally expressed an attitude of cooperation, but added that the effort was a work in progress and talks were ongoing.

The push reflects a widening international debate over how to govern autonomous AI agents as governments increasingly depend on a small number of private companies for the systems that run their digital infrastructure.