Unauthorized group has gained entry to Anthropic’s unique cyber software Mythos, report claims

A gaggle of unauthorized customers has reportedly gained entry to Mythos, the cybersecurity software lately introduced by Anthropic.

A lot has been fabricated from Mythos and its purported energy — an AI product designed for enterprise safety that, within the incorrect palms, may turn out to be a potent hacking software, based on the corporate. Now Bloomberg has reported {that a} “personal on-line discussion board,” the members of which haven’t been publicly recognized, has managed to achieve entry to the software via a third-party vendor.

“We’re investigating a report claiming unauthorized entry to Claude Mythos Preview via certainly one of our third-party vendor environments,” an Anthropic spokesperson advised TechCrunch. The corporate mentioned that, to this point, it has discovered no proof that the supposedly unauthorized exercise has impacted Anthropic’s methods in any method.

The unauthorized group tried a variety of totally different methods to achieve entry to the mannequin, together with utilizing “entry” loved by the one that was interviewed by Bloomberg. That particular person is at the moment employed at a third-party contractor that works for Anthropic, the outlet reported.

Members of the group are a part of a Discord channel that seeks out details about unreleased AI fashions, the outlet reported. The group has been utilizing Mythos usually since getting access to it, and offered proof to Bloomberg within the type of screenshots and a stay demonstration of the software program.

Bloomberg studies that the group, which supposedly gained entry to the software on the identical day it was publicly introduced, “made an informed guess concerning the mannequin’s on-line location primarily based on data concerning the format Anthropic has used for different fashions.” The group in query is “occupied with enjoying round with new fashions, not wreaking havoc with them,” the supply advised the outlet.

Mythos was launched to a choose variety of distributors, together with huge names like Apple, as a part of an initiative known as Undertaking Glasswing. The restricted launch of the mannequin was designed to stop its use by unhealthy actors. The software may very well be weaponized towards company safety as an alternative of bolstering it, Anthropic mentioned.

If true, unauthorized use of Mythos may spell bother for Anthropic, which offered the unique launch to allay the corporate’s concern for enterprise safety.