Google: AI helped fix more Chrome bugs in June than in 2 years

Google says it patched 1,072 security bugs in Chrome in June — more than it fixed across the previous 23 versions released over the past two years combined, which totaled 1,036 fixes. The company credits internal AI tools for the jump.

The count spans the two Chrome versions released in June, Chrome 149 and 150 — what Google calls “milestones.” For comparison, Chrome 126 was released back in June 2024, giving a sense of how much the pace of bug-fixing has accelerated in just two years.

Ever since the advent of large language models, cybersecurity experts have warned that AI-powered systems would let attackers find bugs at an increasingly and exponentially larger scale, forcing defenders to adopt AI of their own to keep up. That prediction is starting to play out, backed by data Google published this week as part of a white paper detailing its use of AI to find and patch flaws faster; a chart in the paper shows the exponential increase in bugs fixed per version.

“By applying models like Gemini, we are preemptively fixing vulnerabilities, outpacing our adversaries and making Chrome safer with every update,” Doug Turner, Chrome’s director of engineering, told TechCrunch in a statement. He said LLMs have fundamentally shifted the economics of cybersecurity, turning vulnerability discovery into an automated, industrial-scale operation.

Google isn’t alone in seeing the trend. Earlier this month, Microsoft said it had patched a record 570 security flaws across its product lines as part of its monthly “Patch Tuesday” release, also citing its own use of AI for the jump.

Apple, however, does not appear to be registering the same exponential increase. An independent count found Apple patched 482 bugs in its products in 2026 — roughly on pace to equal or surpass last year’s total, and also roughly in line with the number of bugs Apple patched back in 2015. TechCrunch reached out to Apple for comment but did not hear back.

Why it matters

If AI tools are making bug discovery and patching an industrial-scale operation for defenders, that same shift almost certainly cuts both ways — the tools available to find flaws are getting more powerful for attackers too, which is exactly what raised the original warning from security researchers.