Google Catches First AI-Generated Zero-Day Exploit

Google’s Threat Intelligence Group (GTIG) has confirmed what cybersecurity experts have been warning about for years: hackers are now using AI to build real, working exploits for previously unknown software vulnerabilities.

In a report published yesterday, GTIG revealed that it caught a threat actor using a zero-day exploit that was almost certainly generated by an AI model. It marks the first documented case of AI being weaponised this way in the wild.

What Happened

The exploit is a Python script designed to bypass two-factor authentication on a widely used open-source web administration tool. The criminal group behind it planned to use the vulnerability in what Google described as a “mass exploitation event” — a coordinated attack targeting many systems at once.

Google’s early discovery likely prevented the attack from happening. GTIG worked with the affected software vendor to patch the flaw before it could be exploited at scale.

How Google Knew AI Was Involved

The code itself gave it away. The Python script was full of overly educational comments, a hallucinated severity score that doesn’t exist in any database, and a textbook-clean structure that’s characteristic of large language model output. Google said it has “high confidence” an AI model was used to both find the vulnerability and write the exploit code.

The company confirmed its own Gemini models were not involved, but noted that threat actors are actively experimenting with agentic AI tools like OpenClaw to automate vulnerability discovery.

A Turning Point for AI Security

This isn’t just a cybersecurity story — it’s an AI story. The same capabilities that make AI tools useful for coding, research, and productivity also make them useful for finding and exploiting software flaws. Google’s report also flagged groups linked to China and North Korea as showing “significant interest” in AI-assisted hacking.

What This Means for You

If you use AI tools in your work, this story is a reminder that AI is a dual-use technology. The tools that help you write code and automate tasks can also be turned against the software you rely on.

For everyday users, the practical takeaway is straightforward: keep your software updated, use two-factor authentication (it’s still far better than passwords alone), and stay informed about how AI is changing the security landscape.

For a deeper look at how AI agents work and why they’re reshaping both productivity and risk, check out our getting started guide or subscribe to our newsletter for daily updates.