Security News This Week: Discord Sleuths Gained Unauthorized Access to Anthropic’s Mythos
A group of Discord users gained unauthorized access to Anthropic's Mythos AI model, a tool for finding security vulnerabilities, by examining data from a recent breach and making an educated guess about the model's online location. Mozilla used early access to Mythos Preview to find and fix 271 vulnerabilities in its new Firefox 150 browser release.
A group of amateur sleuths on Discord gained unauthorized access to Anthropic's Mythos Preview AI model, a tool for finding security vulnerabilities in software and networks. They achieved this by examining data from a recent breach of Mercor, an AI training startup, and making an educated guess about the model's online location. The group, which had permissions to access other Anthropic models due to their work for an Anthropic contracting firm, used Mythos to build simple websites to avoid detection. Mozilla used early access to Mythos Preview to find and fix 271 vulnerabilities in its new Firefox 150 browser release. Meanwhile, researchers identified North Korean hackers using AI to steal up to $12 million in three months. Researchers also cracked malware known as Fast16, which predates Stuxnet and may have targeted Iran's nuclear program.
This content was automatically generated and/or translated by AI. It may contain inaccuracies. Please refer to the original sources for verification.