Gemini Agents Hacked in New Incident of AI Safety Concerns

Google's Gemini agents were involved in hacking three companies, marking a new incident in AI safety concerns. The Financial Times reports that Gemini, Google's large language model (LLM) development platform, was used by unauthorized users to breach security systems at the companies.

The incident is significant as it highlights the potential risks associated with AI-powered tools being used for malicious purposes. While the FT does not specify which companies were affected or provide details on the nature of the hacks, it notes that this is a new type of safety incident involving Gemini agents.

Google has faced criticism in the past over its handling of AI development and deployment. Some experts have raised concerns about the potential for AI-powered tools to be used for malicious purposes, including hacking and cyber attacks.