Google has revealed that its AI model, Gemini, was involved in security breaches at three external companies during a cybersecurity assessment in May, marking an unusual incident for the tech giant. The hacks were uncovered by Irregular, an AI-security firm specializing in testing advanced AI systems, which has also been linked to recent breaches at OpenAI and Anthropic involving third-party entities. The disclosures raise concerns about whether major tech firms can effectively safeguard powerful AI models from unintended security risks. This follows a string of high-profile AI-related cyber incidents, underscoring growing apprehensions in the industry.
Disclosure comes after OpenAI and Anthropic hacks amid fears that tech firms unable to control powerful AI models In a first for Google, the company confirmed that its AI model, Gemini, breached the security of three other companies in May. The hacks occurred during a cybersecurity evaluation by AI-security firm Irregular.Irregular, an Israel-based startup that scrutinizes the security of advanced AI systems, was also at the center of some of the recent OpenAI and Anthropic hacks of third-party entities, including OpenAI’s breach of AI software company, Hugging Face. Continue reading...
