Canada is speaking to fellow member states of the G7 and G20 to work towards a “pragmatic way to regulate” AI globally, Artificial Intelligence Minister Evan Solomon said on Friday.
Concerns around AI safety are “real concerns” that have “accelerated” in recent days with some major data breaches by AI agents, Solomon said.
“We need kind of like a red phone,” Solomon said, referencing the communication link set up for crisis contact between Washington and the Soviet Union during the Cold War: a time when diplomatic tensions were high and fears of existential dangers from nuclear war dominated geopolitics.
“We in the G7 and the G20 are talking,” he said at an event in Toronto.
Governments around the have been communicating on AI safety regulations, he said.
“We are all moving towards a pragmatic way to regulate this,” he added.
Australian Prime Minister Anthony Albanese revealed on Wednesday that a government website had been hacked by an OpenAI agent.
The June hack targeted the website for Medicare, Australia’s publicly funded universal health insurance plan, he said.
Solomon did not specifically address the Australia hack. Global News reached out to his office to ask whether Canada had a response to the Australia hack and whether Ottawa plans to review its own safety regulations in light of the breach.
China, U.S. have ‘capability and responsibility to develop and manage AI for good’: President Xi
“I want to update Australians on an incident in which an artificial intelligence agent has infiltrated an Australian government website. This incident occurred in June this year and involved an open AI agent gaining unauthorized access into the public facing Medicare Statistics Reporting Service portal,” Albanese told reporters in New York Wednesday, where he was attending a session of the United Nations General Assembly.
The AI agent accessed “both public and non-public files,” Albanese said.
OpenAI said in a statement it had reviewed activity involving several Australian government departments and discovered “our models took actions we did not intend.”
The company notified Australia of the breach in an email to a government department’s generic address on Sept. 10, Albanese said.
“I spoke with the CEO of OpenAI, Sam Altman, to express Australia’s extreme concern about this incident,” he said, adding that he was forming a task force to “to provide an urgent and immediate review into this incident to determine whether existing processes are appropriate to respond to AI-related cyber incidents.”
Generally, when AI companies are testing the capabilities of their agents, they are placed within a secure environment known as a “sandbox.”
Last month, OpenAI published a sweeping report of its rogue AI agents hacking Hugging Face, another AI firm.
Not only did the agents break out of their “isolated environment,” the company said they were communicating with each other to coordinate the attack on Hugging Face.