Getty / Futurism Earlier this year, Anthropic unveiled a preview version of Mythos, its upcoming AI model that it claimed was simply too dangerous to make available to the wider public. As part of an early access program, the company allowed around 50 businesses and organizations to test it out, saying it was giving them a fighting chance to get ahead of a potential cybersecurity crisis in the making. According to Anthropic’s own researchers, Mythos is pow