Google says Chrome AI bug fixes topped two years of patches in June

Google says Chrome AI bug fixes surged in June, with its internal tools helping engineers patch 1,072 security flaws across two browser releases. For everyday users, that means Chrome received an unusually large security cleanup, and for Big Tech watchers, it is a concrete example of AI being used inside a core product rather than just sold as a feature.

Google announced Thursday that the fixes appeared in Chrome 149 and Chrome 150, both released in June. The company said that total exceeded the 1,036 security fixes made across the prior 23 Chrome versions over the past two years.

A security bug is a flaw that could let attackers abuse software in ways the developer did not intend. Patching means changing the code to close that weakness, usually through a software update users receive through the browser’s normal release process.

How did Google fix so many Chrome bugs with AI?

Google said the increase came from using internal AI tools to find and repair vulnerabilities faster. In a white paper on its security work, the company published a chart showing a steep rise in fixed Chrome security bugs, using the company’s term “milestones” for Chrome versions.

Chrome 126 was released in June 2024, according to Google’s developer release notes. The latest two milestones cited by Google, Chrome 149 and Chrome 150, were released in June 2026.

Doug Turner, Chrome’s director of engineering, told TechCrunch that large language models have changed the cost and scale of finding vulnerabilities. Turner said Google is applying models such as Gemini to fix weaknesses earlier and make Chrome safer through updates.

Large language models, or LLMs, are AI systems trained to process and generate language and code-like text. In security work, companies can use them to scan software, spot suspicious patterns, help write fixes, or speed up the review process, although the final workflow can vary by company.

Microsoft has reported a similar jump

Google is not the only major tech company tying a rise in security fixes to AI. Earlier in July, Microsoft said it patched a record 570 security flaws across its products in its monthly scheduled security update cycle, commonly called Patch Tuesday.

Microsoft also pointed to its use of AI as a reason for the higher number of fixes, according to TechCrunch. The pattern supports warnings from cybersecurity experts that AI tools could sharply increase how many flaws are found, forcing software makers to use similar tools to stay ahead of attackers.

Apple does not appear to be showing the same kind of spike, based on the figures cited by TechCrunch. An independent count of Apple product bug fixes showed 482 patches in 2026, a pace roughly in line with matching or beating last year’s total and similar to the number Apple patched in 2015.

TechCrunch said it asked Apple for comment and did not receive a response.

The big question for users is practical: more fixes can mean more discovered risk, but it can also mean companies are finding and closing holes faster. Google’s numbers show AI is already changing the volume of security work behind widely used software like Chrome.

This story draws on original reporting from TechCrunch.