Anthropic has been using the Claude Mythos model internally since February 24th, 2026.
AI Fact-Check
“Multiple sources confirm that Anthropic developed a new frontier model called Claude Mythos Preview, which is not publicly available. An article from GLN-7.5 published on April 9, 2026, explicitly states, "Anthropic has been using Mythos internally since February 24th, 2026." While official Anthropic announcements from early April 2026 confirm the model's existence and internal deployment, they do not specify the exact start date. However, the specificity of the date in a secondary source that accurately reports other verified details lends strong support to the claim. Context: On April 8, 2026, Anthropic announced Project Glasswing, a cybersecurity initiative giving select partners access to Claude Mythos Preview. The model was deemed too powerful for public release due to its advanced ability to find and exploit software vulnerabilities.”
Source Videos (1)
Claude Mythos and the end of software
Theo - t3․gg
Related Claims
Anthropic's Claude Mythos model is a much bigger, more expensive, slower, but more powerful model compared to Opus.
Anthropic released a new AI model named Mythos, which is an upgrade from previous models like Sonnet, Opus, and Haiku.
Anthropic is creating entire features in days, not weeks, in early 2026.
Anthropic will not widely release its Claude Mythos Preview model because of its potential to cause harm if misused.
In early internal use, Claude Mythos preview reached unprecedented levels of reliability and alignment, leading to broader use with less frequent human interaction than prior models.