Skip to content
AI IntelligenceMay 16, 2026AI Intelligence
Article

Researchers at Carnegie Mellon University built a new benchmark showing that AI agents like Claude Mythos and GPT-5.5 can...

Autonomously exploit real browser vulnerabilities. Mythos leads GPT-5.5 by a wide margin but costs twelve times as much.

Data Cube AI EditorialSource: The Decoder
01

Source Brief

Researchers at Carnegie Mellon University built a new benchmark showing that AI agents like Claude Mythos and GPT-5.5 can autonomously exploit real browser vulnerabilities. Mythos leads GPT-5.5 by a wide margin but costs twelve times as much.