Anthropic is expanding access to its AI tool, Mythos, by allowing 150 additional organizations globally, bringing the total number of groups with access to around 200. This expansion includes entities from vital infrastructure sectors such as power, water, healthcare, and communications, aimed at enhancing their cybersecurity capabilities. Although Anthropic has labeled Mythos as dangerous due to its potential for offensive cybersecurity actions, it emphasizes that the tool’s primary use is for defensive purposes, particularly through its Project Glasswing, which focuses on securing critical software systems.

Mythos: Mythos refers to Anthropic’s advanced AI model preview, noted for its exceptional capabilities in identifying and addressing software vulnerabilities. It is central to the news as the company widens access to the model for select organizations despite highlighting its potential risks in cyber operations.
Anthropic: Anthropic is an artificial intelligence company focused on developing advanced language models under the Claude family. In the context of this news, the company is actively expanding controlled access to its specialized Mythos model through the Project Glasswing initiative to enhance cybersecurity across critical sectors worldwide.

`json
{
“Risk Assessment”: “Anthropic has stated the potential risks associated with Mythos, given its powerful cybersecurity capabilities aimed primarily at bolstering defensive measures.”,
“Global Expansion”: “The expansion includes organizations from multiple countries focused on essential infrastructure sectors such as power, water, healthcare, and communications.”,
“Cybersecurity Initiative”: “Anthropic’s Project Glasswing deploys Mythos specifically to help partners scan and secure critical software systems.”
}
`