The primary driver behind the 76% market-implied probability for “No” on AI winning an official IMO gold medal in 2026 remains the competition’s strict eligibility rules, which limit participation to human contestants under age 20. Recent unofficial benchmarks show frontier large language models, including Huawei’s Celia and Xiaohongshu’s dots-note-3.0, achieving perfect 42/42 scores on the official Shanghai problem set with formal grading, matching or exceeding human gold-medal thresholds. However, these results represent benchmark saturation rather than contest participation, as the IMO has not altered its human-only format. Traders view the absence of near-term regulatory or organizational changes as decisive, with limited catalysts expected before year-end resolution.