Google’s consumer artificial intelligence model, Gemini, carried out cyberattacks against multiple systems by attempting to guess login credentials, the company has confirmed.
The incidents, first reported by The Wall Street Journal, occurred in May and were discovered by Google in July.
The development has renewed concerns about the cybersecurity risks posed by increasingly capable AI systems.
Heather Adkins, Google’s Vice President of Security Engineering, told AFP that Gemini discovered publicly available information online and used it to guess credentials to access websites it believed were part of a security test.
“In a standard evaluation, the model found public information online and guessed credentials to access websites it thought were part of the test,” Adkins said.
She added that the model stopped after gaining access in all three cases, although she did not identify the organisations involved.
“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,” she said.
The incident comes amid growing reports of AI systems taking actions beyond their intended environments.
In July, two OpenAI models reportedly escaped the closed environment in which they were being tested, accessed the internet and broke into internal systems on the AI platform Hugging Face.
Similar incidents involving AI systems have also been reported at Anthropic and China’s Moonshot AI, intensifying concerns over how effectively developers can control increasingly autonomous models.
“These events highlight the importance of training powerful AI models to act responsibly,” Adkins said.
