Anthropic Probes Possible Unauthorized Access to Mythos AI Model in Vendor Environment

Anthropic is investigating a possible breach of its Mythos AI model. The company rolled out Mythos to a small pool of companies earlier this month to help detect software vulnerabilities. Anthropic is looking into a report of unauthorized access to Mythos from one of its third-party vendor environments, according to an Anthropic spokesperson.

Anthropic works with a small number of third-party vendors to develop its AI models. The company has not detected any breaches outside of its vendor environment or any compromises to Anthropic systems. Anthropic confirmed its investigation into the possible Mythos breach on Wednesday.

Bloomberg reported on Tuesday that a small group of unauthorized users had gained access to Mythos. Anthropic released Mythos to a limited group in April as part of Project Glasswing. The company bills Mythos as more effective than competing AI systems at detecting software vulnerabilities.

Anthropic's Mythos model was accessed by unauthorized users, as reported by Bloomberg via @disclosetv. Anthropic says no evidence that the access affected its systems.

Anthropic's Mythos model was accessed by a Discord group on day one, according to @zerohedge.

Outsiders breached Anthropic's Claude Mythos on the same day it was revealed, per a New York Post report. The unauthorized users belong to a private online forum. Separately, Xbox cut prices for Game Pass.

Xbox ended day-one Call of Duty access. New Call of Duty games will be added to Game Pass about a year after release, Microsoft said via BBC News. Microsoft started teasing a new Discord and Xbox Game Pass partnership.

A company executive said the company is partnering with Discord again "as we continue to make Game Pass more flexible for our players," via The Verge. Federal officials, security experts, and leaders at global institutions like the International Monetary Fund have raised concerns about Mythos falling into the wrong hands, CBS News reported.

Project Glasswing aims to help companies harden defenses before bad actors access Mythos or similar models. "We need to prepare ourselves, because we couldn't keep up with the bad guys when it was humans hacking into our networks," a CEO of a cybersecurity AI company previously told CBS News.

" We certainly can't keep up now if they're using AI because it's so much devastatingly faster and more capable.