LLMs kill the 90-day patch window with faster exploit cycles
- Source
- Himanshu Anand
- Time
- 8:47 AM
- Weight
- 95/100
The traditional 90-day responsible disclosure window is increasingly viewed as obsolete due to the rapid integration of Large Language Models (LLMs) in cybersecurity. Industry experts argue that AI-powered tools have significantly compressed the time required to discover vulnerabilities and develop working exploits.
Recent incidents, such as the discovery of the "Copy Fail" Linux kernel flaw, demonstrate that AI can identify complex logic errors in minutes—tasks that previously required weeks of manual labor by highly skilled researchers. This acceleration has shifted the threat landscape, as multiple independent researchers and malicious actors now frequently converge on the same bugs simultaneously.