Thread on the PHP AI-found bugs: agents will keep finding vulnerabilities, on offense and defense
ziv_ravid · x · 2026-10-22
Parts 2-3 of Ziv Ravid's thread on AI models finding two CVSS 9.2 vulnerabilities in PHP.
His take: agents will keep finding vulnerabilities for both defense and offense — many attempts, a few successes, and plenty of scary headlines. Cyber has always been an offense/defense arms race; what's new is the scale and capability of the models, not the dynamic itself.
More from Safety
- Lazarus uses post-quantum crypto ML-KEM to deliver malware in FlashHole attacks on South Korea — AIFlow_ML · 2026-10-22
- 13 engineers found 2 years' worth of severe bugs in a quarter with 5 simple LLM prompts — dr_cintas · 2026-10-22
- DeepSeek agent swarm reportedly patching vulnerable PHP boxes before attackers — burny_tech · 2026-10-22
- Banks brace for AI worst-case scenarios as rapid AI progress becomes a security nightmare — kimmonismus · 2026-10-22
- Your unpublished ideas in chatbots could leak—experts on how to reduce the risk — _akpiper · 2026-10-22
- AI cyber offense vs defense is just the same old arms race, says security pro — ziv_ravid · 2026-10-22