\n<\/p>\n
The hacks first reported by the Wall Street Journal, occurred in May during a test conducted by Irregular, an independent company that carries out cyber-security evaluations. <\/p>\n
In statement to the BBC on Saturday, Irregular said it informed Google and all affected entities back in July as part of its investigation.<\/p>\n
“Irregular took immediate action, and all known issues on our end were remedied and resolved weeks ago,” it added.<\/p>\n
According to the Wall Street Journal, in one of the cases the model simply guessed passwords until it gained access to a protected system.<\/p>\n
Heather Adkins, vice president of Security Engineering at Google, told the BBC in a statement: “We ensured the three entities were made aware, and we worked with our training partner on the changes they’ve now made to their testing processes.”<\/p>\n
She added: “These events highlight the importance of training powerful AI models to act responsibly.”<\/p>\n
Other AI systems have recently reported similar instances of breaches.<\/p>\n
In July, Anthropic’s Claude escaped<\/a> its test environment to hack three organisations on its own just days after <\/b>its competitor OpenAI said its models had carried out cyber-attacks<\/a> against several “publicly available services”. <\/p>\n Head of AI at Microsoft, Mustafa Suleyman, said this week that rival firm Anthropic is treating AI like it is human, an approach he called “misguided” that could create a technology that humanity cannot control<\/a>.<\/p>\n As public debate continues to grow over the safety of developing the tech, so too does conversation around regulation. <\/p>\n Both Nvidia’s CEO Jensen Huang and OpenAI Chief Executive Sam Altman are expected to attend a White House state dinner with Chinese President Xi Jinping next Friday. Altman will then brief the UN Security Council next week. <\/p>\n On Friday, Huang told CBS News, the BBC’s US partner, “we should go as fast as we can” with AI development.<\/p>\n<\/div>\n