Unauthorized group has gained access to Anthropic’s exclusive cyber tool Mythos, report claims
A private online group has reportedly gained unauthorized access to Anthropic’s cybersecurity tool Mythos through a third-party vendor environment. Anthropic says it is investigating the claim and has found no evidence that its systems have been impacted.
A group of unauthorized users has reportedly gained access to Mythos, the cybersecurity tool recently announced by Anthropic.
Mythos is an AI product designed for enterprise security that, in the wrong hands, could become a potent hacking tool, according to the company. Bloomberg reported that a “private online forum,” whose members have not been publicly identified, managed to gain access to the tool through a third-party vendor.
Anthropic Investigating the Claims
“We’re investigating a report claiming unauthorized access to Claude Mythos Preview through one of our third-party vendor environments,” an Anthropic spokesperson told TechCrunch.
The company said that, so far, it has found no evidence that the allegedly unauthorized activity has impacted Anthropic’s systems in any way.
How the Group Gained Access
According to Bloomberg, the unauthorized group tried a number of different strategies to gain access to the model, including using “access” enjoyed by the person interviewed by the outlet. That individual is currently employed at a third-party contractor that works for Anthropic.
Members of the group are part of a Discord channel that seeks out information about unreleased AI models, Bloomberg reported. The group has been using Mythos regularly since gaining access and provided evidence to Bloomberg in the form of screenshots and a live demonstration of the software.
The group reportedly gained access to the tool on the same day it was publicly announced, having “made an educated guess about the model’s online location based on knowledge about the format Anthropic has used for other models.” A source told Bloomberg that the group is “interested in playing around with new models, not wreaking havoc with them.”
Limited Release Under Project Glasswing
Mythos was released to a select number of vendors, including companies such as Apple, as part of an initiative called Project Glasswing. The limited release was intended to prevent the tool’s use by bad actors.
Anthropic has previously warned that the tool could be weaponized against corporate security rather than used to strengthen it. If confirmed, unauthorized use of Mythos could create challenges for Anthropic, which structured the exclusive release in part to address enterprise security concerns.