3 hours ago
10
OpenAI has broadened access to its GPT 5.4 Cyber model, countering Anthropic’s Mythos reveal under Project Glasswing. Anthropic’s ranking as the third best AI model by April 2026 is at 28% YES, up 15% since the announcement.
The expansion of GPT 5.4 Cyber’s user base to more vetted defenders has shifted trading in AI model ranking markets. Anthropic’s Claude Mythos is now priced as a stronger contender for the third best AI model spot. With 14 days until April 30, the 15-point jump suggests traders expect public benchmark results to validate the model’s capabilities, even with limited access so far.
Volume analysis suggests this isn’t purely speculative. Face value trades are at $0, and the rise is likely driven by strategic positioning ahead of potential benchmark results. This pattern matches historical market behavior where positioning precedes official evaluations or announcements. That said, without actual trade data, the odds shift may reflect perception rather than new substantive information.
Anthropic is priced as a serious competitor against OpenAI and Google, and that perception supports the likelihood of Claude Mythos being validated by public benchmarks. At 28¢, a YES share pays $1 if Mythos ranks third by the end of April, a potential 3.6x return. For this bet to pay off, traders need Mythos to post benchmark results within two weeks.
Watch for LMSYS Chatbot Arena benchmark updates and any leaks from the Project Glasswing consortium. Either could move Anthropic’s odds sharply.
Get prediction market intelligence as a structured API feed. Early access waitlist.
English (US) ·