Anthropic Accidentally Leaks Next-Generation Model 'Claude Mythos': A New Level Beyond Opus with Cybersecurity Capabilities 'Far Superior to All Existing AIs'

According to monitoring by 1M AI News, Anthropic is testing a new model named ‘Claude Mythos’. Due to a misconfiguration in their content management system, an unpublished blog draft was stored in a publicly searchable database, inadvertently exposing the existence of this model. Anthropic subsequently confirmed the model, stating it represents a ‘step change’ in capabilities and is ‘the strongest model we have built to date’. The leaked draft indicates that Mythos is a completely new model tier, positioned above the Opus series. The draft claims that ‘compared to the previous best model Claude Opus 4.6, this model shows significant improvements in software programming, academic reasoning, and cybersecurity tests’. The draft also mentions the codename ‘Capybara’, which points to the same underlying model as Mythos. Anthropic is particularly cautious about the model’s cybersecurity capabilities, stating that it ‘currently far outpaces any other AI model in terms of cybersecurity capabilities’ and ‘signals an impending wave of models whose exploit capabilities will far exceed the defenders’ response speed’. As a result, Anthropic has adopted a slower release strategy than before: initially opening it to a small number of early customers, focusing on cybersecurity defense scenarios to help organizations fortify their codebases before the onslaught of AI-driven attacks. The operational costs of this model are extremely high, and it will not be made publicly available in the short term. The leak originated from a ‘human error’ in Anthropic’s external CMS tool, resulting in nearly 3,000 unpublished assets (including blog drafts, images, and PDFs) being publicly stored in an unencrypted database. Cybersecurity researcher Alexandre Pauwels from Cambridge University and Roy Paz, a senior AI security researcher at LayerX Security, independently discovered these files. The leaked content also includes details about an upcoming closed-door summit for European CEOs that Anthropic will host in the UK, which CEO Dario Amodei will attend at an 18th-century manor hotel in the English countryside. An Anthropic spokesperson stated: ‘We are developing a general model that has made significant progress in reasoning, programming, and cybersecurity. Given the strength of its capabilities, we will be very cautious in our release approach.’