Forget the AI Slowdown—the Vulnerability Explosion Is Already Happening (www.wired.com)

A recent surge in software vulnerabilities has been uncovered through the use of AI tools, raising significant alarms within the AI and cybersecurity communities. Despite ongoing discussions about slowing down AI development to avert extreme scenarios, the reality is that AI-enabled bug hunting has already led to an unprecedented explosion of vulnerabilities. For instance, Microsoft recently reported patching 974 Common Vulnerabilities and Exposures (CVEs) in a single month—a new record—while other tech giants like Oracle and Google are also experiencing sharp increases in vulnerability disclosures. The total number of CVEs has skyrocketed to over 66,000 this week, revealing a dramatic uptick in security flaws compared to previous years. This trend underscores a crucial challenge: while the ability of AI to discover vulnerabilities has improved the overall visibility of security flaws, the pace of patching and remediation has not kept up, potentially leading to a surge in cyberattacks from malicious actors also leveraging AI. Experts remain divided on the implications of this spike, with some asserting that merely increasing the number of known vulnerabilities does not equate to heightened risk, as it is a sign of systems functioning properly. Nevertheless, there is a consensus that the overwhelming number of vulnerabilities could strain already resource-limited IT teams and exacerbate existing cybersecurity challenges, prompting a critical dialogue on how to balance the benefits of AI in security with the risks it introduces.