Google said Thursday it patched 1,072 security bugs across the two latest versions of Chrome, both released in June, more than the 1,036 bugs fixed across the previous 23 versions released over the prior two years combined. The company credits its internal AI tools, built on its Gemini models, for the jump. A single month’s bug count exceeding two full years of prior fixes is what NewsTrackerToday traces to as the more startling number than the AI-driven explanation behind it.
Chrome’s director of engineering, Doug Turner, framed the shift in explicitly economic terms: “LLMs have fundamentally shifted the economics of cybersecurity, transforming vulnerability discovery into an automated, industrial-scale operation. By applying models like Gemini, we are preemptively fixing vulnerabilities, outpacing our adversaries and making Chrome safer with every update.” That framing treats bug discovery itself as a now-automatable, scalable process, a meaningfully different description of security work than the manual, researcher-driven model the industry has relied on for decades.
Sophie Leclerc, who covers the technology sector, reads the timeline embedded in Google’s own chart as the clearest evidence of the shift: “Chrome 126 shipped in June 2024 with a modest bug count typical of that era. Two years later, Chrome 149 and 150 alone produced more fixes than the 23 versions in between combined. That’s not a gradual improvement curve, it’s a step change coinciding almost exactly with when AI-assisted vulnerability discovery tools matured enough for large engineering teams to deploy them at scale.” That step-change timing, more than the raw bug count, is what NewsTrackerToday folds on as the more precise evidence AI tools are doing real, measurable work here.
Google isn’t alone in seeing this pattern: Microsoft announced earlier this month that it patched a record 570 security flaws across its product lines in a single monthly release cycle, also crediting its own AI tools for the jump, suggesting this isn’t a Chrome-specific phenomenon but a broader shift across major software vendors using AI to accelerate vulnerability discovery.
Daniel Wu, who covers geopolitics and energy, reads the notable exception as the more revealing data point: “Apple, by an independent count, patched 482 bugs in 2026, roughly on pace with its 2025 total and even its 2015 total, showing no comparable exponential jump. Apple didn’t respond to requests for comment on why it isn’t seeing the same acceleration Google and Microsoft are reporting. That gap raises a real question about whether Apple’s own AI-assisted security tooling lags behind its two biggest rivals, or whether Apple’s development process is simply structured differently in ways that don’t lend themselves to the same automated discovery approach.” That unexplained gap, more than Google’s own numbers, is what News Tracker Today hinges round as the more interesting open question in this week’s disclosures.
Cybersecurity researchers have warned for roughly two years that AI-powered tools would eventually find vulnerabilities at an exponentially increasing rate, forcing defenders to adopt the same AI tools just to keep pace with attackers capable of the same automated discovery techniques, a prediction Google’s own data now appears to confirm directly.
None of this confirms whether this dramatic increase in discovered and patched bugs reflects genuinely more vulnerabilities existing in Chrome’s code, or simply far better tools for finding vulnerabilities that were always present but previously went undetected. Whether this AI-driven bug-hunting pace holds steady, accelerates further, or eventually plateaus once the most findable vulnerabilities have been surfaced, and whether Apple’s tooling eventually catches up to match it, is what NewsTrackerToday lands to as the real question this month’s numbers leave open.
