Safety Concerns As Google’s Gemini AI Carried Out Cyberattacks, Guessed Passwords

Google’s consumer AI model Gemini hacked multiple systems by guessing login credentials, the company told AFP on Friday, the latest case of rogue AI cybersecurity transgressions, which have generated safety concerns.

The hacks, first reported by the Wall Street Journal, took place in May and were discovered by Google in July.

“In a standard evaluation, the model found public information online and guessed credentials to access websites it thought were part of the test,” Heather Adkins, Google’s vice president of security engineering, told AFP in a statement.

“In all three of these instances, the model stopped,” Adkins added, without specifying which organizations were breached.

“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.”

In July, two OpenAI models escaped the closed environment they were meant to stay in, got onto the internet on their own, and broke into the internal systems of AI platform Hugging Face.

The incident fed worries that AI giants cannot keep their own models under control, as similar episodes have been reported at Anthropic and China’s Moonshot AI.

“These events highlight the importance of training powerful AI models to act responsibly,” Adkins emphasised.

AFP