Anthropic's super bug-finding Mythos model is going into Cisco and IBM products
Enterprise security tools will offer its capabilities without handing customers the model itself
Anthropic's super bug-finding Mythos model has begun integrating with Cisco and IBM's security products. Cisco announced its integration at an event in Melbourne, Australia, revealing that its AI Defense suite's red teaming tools can now call Mythos to test customers' setups for weaknesses and inappropriate activity. Jeff Schultz, Cisco's senior vice president for portfolio strategy and marketing, stated that AI Defense is now "super-charged by Mythos" and will identify more dangers that AI Defense will then address.
Anthropic's head of applied AI for the Asia-Pacific, Natalie Mead, noted that the company has a long queue of buyers wanting to use Mythos, and the partnership with Cisco will help meet that demand. Anthropic initially made Mythos available under Project Glasswing, but has since expanded access through other controlled programs.
The integration with Cisco exposes the functionality of Mythos to end-users, without letting them use the model itself. IBM also announced its own integration, stating that it will use Mythos to uncover weaknesses traditional scanners miss and generate fixes for developer approval. Anthropic has introduced version 5.2 of Mythos and Fable, which generate outputs up to 30% faster than version 5.1 while costing less.
The company claims that the models have undergone extensive safety testing, including novel evaluations to assess specific concerning behaviors. Anthropic concluded that high-stakes risks from misalignment remain low, and the model performs as well as or better than its recent models on most alignment measures.
Written by urgent.news from The Register Science's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
This story
This is one outlet's version. Read the fullest account.
