Large Language Models
Apr 20, 2026
Anthropic's Mythos AI Model Limited to Select Businesses for Cybersecurity Testing
Apr 20, 2026
AI Summary
Anthropic has developed a new AI model, Mythos, designed to identify and exploit cybersecurity vulnerabilities. The company has chosen to restrict its release to a limited number of businesses to allow for testing and to enhance software security against potential AI-driven attacks.
- Anthropic, led by Dario Amodei, announced the creation of the Mythos AI model, which excels in identifying cybersecurity vulnerabilities.
- The company decided not to release Mythos to the public, opting instead to provide access to a select group of businesses for testing purposes.
- This decision aims to help these organizations strengthen their software defenses against possible AI-enabled cyber threats.
- The announcement has raised concerns among leaders in both business and government sectors.
- A Live Q&A session was held on April 17, 2026, featuring discussions among Bloomberg Tech co-host Edward Ludlow and reporters Shirin Ghaffary, Margi Murphy, and Todd Gillespie.
anthropicmythosqaai modellanguage model