Google Warns of AI-Generated Zero-Day Exploits Used by Hackers

Google's Threat Intelligence Group has reported a significant escalation in the cyber landscape: hackers are now successfully using artificial intelligence to discover and develop zero-day exploits. This development confirms long-standing fears that generative AI could be used to lower the barrier for creating complex malware.

The researchers discovered that a major security flaw was developed with the assistance of AI models, allowing attackers to exploit vulnerabilities before developers could issue patches. This discovery has prompted a surge in investment from security firms—including Google, OpenAI, and Anthropic—as they race to use their own AI models to identify and patch critical network weaknesses before they can be exploited.

The report highlights that we have entered a "dual-use" era where the same intelligence used to protect systems is being mirrored by bad actors to dismantle them.