ICE has adopted an AI system named Claude Mythos through its Project Glasswing initiative to accelerate vulnerability detection, prioritizing precision and rapid remediation over traditional methods. The move underscores a growing trend among major firms to leverage generative AI for defensive cybersecurity tasks, rather than just automation.
Executives detailed that the focus is on improving the accuracy of threat identification and the speed of delivering fixes, though specific metrics or deployment timelines were not disclosed. The effort appears to target internal network security as well as potential client-facing systems, though the scope remains tightly held.
The initiative arrives as mortgage and real estate technology firms face increasing scrutiny over data protection, with ICE handling vast amounts of sensitive financial and property records. Claude Mythos is described as a specialized large language model trained on security-specific datasets, though details on its training data or architecture remain proprietary.
Critics caution that AI-driven cybersecurity tools can introduce new attack surfaces, such as model poisoning or adversarial inputs, and that without transparent testing, the claims of improved speed and accuracy remain unverifiable. ICE has not released independent audit results.