tech
Anthropic limits Mythos AI rollout over fears hackers could use model for cyberattacks
Microsoft, Amazon, Apple, CrowdStrike, Palo Alto Networks and others will use the model as part of a new cybersecurity initiative called Project Glasswing.
TL;DR
- Anthropic unveiled Claude Mythos Preview, an advanced AI model that identifies software weaknesses and security flaws.
- The model is being rolled out through a new cybersecurity initiative called Project Glasswing to prevent misuse by bad actors.
- Initial launch partners include Apple, Google, Microsoft, Nvidia, and Amazon Web Services, along with over 40 other cybersecurity companies.
- The goal of Project Glasswing is to allow cyber defenders to use the model for defensive security work and gain a head start in identifying vulnerabilities.
- Claude Mythos Preview can detect critical bugs that have been difficult to find, including a 27-year-old bug in OpenBSD.
- Anthropic emphasizes responsible AI deployment and is engaging with U.S. government officials on the model's capabilities.