Security leaders warn AI will accelerate vulnerability discovery and exploitation

The fallout from recent critical vulnerabilities in PHP—the language powering roughly 70% of websites—continues even as media attention fades. Organizations are still in cleanup mode after two major bugs forced emergency patching and some firms to temporarily take systems offline.

A coordinated crisis response#

The response to these flaws involved a rapid mobilization across the tech industry. Security researchers at Wiz, alongside teams from OpenAI, the research collective Calif, and the PHP development team itself, worked in concert to identify the vulnerabilities and develop patches before widespread exploitation.

The AI acceleration threat#

Rather than treating this as a contained incident, security experts are pointing to a larger pattern emerging. As artificial intelligence technologies become more sophisticated, the velocity at which new vulnerabilities can be discovered—and subsequently weaponized—is expected to increase substantially. This shift has prompted warnings about what some are calling a coming "vulnpocalypse."

Tomer Schwartz, who leads exposure management and application security efforts at Wiz (owned by Google), recently discussed the implications of this threat on TechCrunch's Equity podcast. Schwartz and host Rebecca Bellan explored how the recent PHP incident unfolded behind the scenes, what strategies emerging companies should adopt to build security-first, and why manual patching approaches cannot scale to meet future demands.

Preparing for the next wave#

The conversation underscores a growing consensus: reactive vulnerability management is becoming inadequate. As AI tools lower the barriers to finding flaws, the industry must shift toward proactive security practices and automated remediation systems that can keep pace with the accelerating threat landscape.

The full episode is available on YouTube, Apple Podcasts, Overcast, Spotify, X, and Threads.