# Anthropic's Secret Mythos AI Model Exposed After Data Leak Reveals Step Change in Capabilities

*2026-03-30 11:01 UTC · Technology · 1 min read*

> Anthropic confirmed testing its most powerful AI model yet, Claude Mythos, after security researchers found leaked internal documents in a public data cache.

Security researchers Roy Paz of LayerX Security and Alexandre Pauwels of the University of Cambridge discovered roughly 3,000 unpublished assets in a publicly accessible Anthropic data cache, including a draft blog post revealing a new AI model called Claude Mythos. Anthropic confirmed the leak resulted from "human error" in its content management system and acknowledged it is testing the model with early access customers.

The leaked documents describe Mythos as a tier above Opus 4.6, with "dramatically higher scores" in coding, academic reasoning, and cybersecurity. Anthropic warned the model is "currently far ahead of any other AI model in cyber capabilities" and "presages an upcoming wave of models that can exploit vulnerabilities in ways that far outpace the efforts of defenders."

Anthropic stated: "We're developing a general purpose model with meaningful advances in reasoning, coding, and cybersecurity. Given the strength of its capabilities, we're being deliberate about how we release it."

## Sources

- [fortune.com](https://fortune.com/2026/03/26/anthropic-says-testing-mythos-powerful-new-ai-model-after-data-leak-reveals-its-existence-step-change-in-capabilities/)
- [techzine.eu](https://www.techzine.eu/news/applications/140017/details-leak-on-anthropics-step-change-mythos-model/)

---
TechTools News · https://techtools.cz/tools/news/article/1153-anthropic-s-secret-mythos-ai-model-exposed-after-data-leak-reveals-step-change
Česká verze: https://techtools.cz/tools/news/clanek/1201-tajny-ai-model-anthropicu-mythos-odhalen-po-uniku-dat-predstavuje-skokovy-posun
