Anthropic has launched its Claude Mythos AI model focused on cybersecurity, alongside a safer version for public use. This move highlights the company's commitment to enhancing AI safety in the tech landscape.