Skip to content
GPT-5.5: Mythos-Like Hacking, Open to All
xbow.com | blog | #ai-security | #benchmark | #offensive-security | #vulnerability-detection | #gpt-5-5 | #open-access
Summary
XBOW's early access to GPT-5.5: the model cuts its vulnerability miss rate to 10% from GPT-5's 40%, a leap in black-box performance that effectively retires its white-box benchmark.
- Published
- Collected