A group of unauthorized users has reportedly gained access to Anthropic's cyber tool, Mythos, which was recently announced as a powerful AI product. The unauthorized group, who have not been identified, reportedly used the same access to the tool on the same day it was publicly announced. An Anthropic spokesperson stated that there is no evidence that this unauthorized activity impacted Anthropic’s systems. The group, part of a Discord channel that seeks information about unreleased AI models, has been using Mythos regularly and provided evidence via screenshots and a live demonstration of the software. If true, this could cause trouble for Anthropic, which provided the exclusive release to several vendors as part of an initiative called Project Glasswing.