Google Reports Surge in Software Security Flaws Driven by AI Automation

Google's threat research team published findings showing that reported software security flaws doubled to over 10,000 per month by August 2026. According to the report covered by Recorded Future News, malicious groups are using automated artificial intelligence tools to study security patches and rapidly build attacks against newly patched systems.

This shift means the window of safety between when a company releases a software update and when criminals exploit the flaw has shrunk dramatically. Instead of spending weeks searching for brand new security holes, attackers can feed software update notes into language models to quickly figure out how to break in. For everyday users and businesses, delaying software updates even by a few days creates a major security hazard because automated tools allow hackers to target unpatched devices on a massive scale.

It remains uncertain how software vendors will keep up with defender side AI to patch flaws faster than attackers can weaponize them. If defensive teams are forced to deploy automated patch updates without thorough testing, it could lead to widespread system crashes or unexpected software breakage. Whether tech companies will eventually restrict AI tools from analyzing public security advisories altogether is another open question for the industry.