Stories

Fri, 17 Apr 2026

Fri, 17 Apr 2026 What is Claude Mythos and what risks does it pose?

The company's claim the AI tool can outperform humans at some hacking and cyber-security tasks has sparked fears in the financial world.

* Anthropic's AI model, Claude Mythos, has been reported to outperform humans at some hacking and cyber-security tasks.
* The company claims the tool can locate dormant bugs in old code and exploit them, but some experts remain skeptical about its capabilities.
* 12 tech companies have been given access to Mythos via Project Glasswing, an initiative to strengthen resilience against the model's potential threats.
* Regulators, legislators, and financial institutions are discussing the dangers of Mythos, with some expressing serious concerns that it could undermine digital security.
* The EU has started discussions with Anthropic about its concerns around Mythos, while the US government has offered to work with Anthropic to "help defend against the risk" of the model.
* Some experts believe that Mythos' capabilities are overhyped and that the tool's biggest threat would be against poorly defended systems.
* The UK's AI Safety Institute recently concluded that while a very powerful model, its biggest threat would be against vulnerable systems.