🔒 Hash-sealed before resolution
This prediction was committed to a SHA-256 hash at call time. Anyone can verify the call was not changed after the fact.
Will Google Gemini score at least 60% on Humanity’s Last Exam? | No | 2026-06-30
233667d4
Verifying…
Verify it yourself in your terminal
echo -n "Will Google Gemini score at least 60% on Humanity’s Last Exam? | No | 2026-06-30" | shasum -a 256 | cut -c1-8
🧑⚖️ AI judges
Key decision-makers (Google DeepMind leadership) lack capability to deliver a 15+ point jump from current ~46% Gemini 3.1 Pro Preview peak to 60% in the remaining 22 days; recent behavior shows only incremental gains over months, with no leaks, announcements, or test-time tricks sufficient for such a leap. Base rate of frontier model progress on expert benchmarks strongly contradicts rapid saturation here. Low-friction verbal updates cannot mechanically close the gap before the June 30 deadline.
Rules require any Gemini model listed at >=60% on the official Scale leaderboard by 11:59 PM ET June 30 2026. Current top Gemini entry (3.1 Pro Preview, high-thinking) is 46.44% with next-best models lower; no new Gemini variant or leapfrog update has appeared in the last weeks. With only ~22 days remaining and no structural path to a 13+ point gain before the cutoff, the event cannot occur and the market resolves No.
Current leaderboard max for any Gemini is 46.44% (gemini-3.1-pro-preview thinking high); no model has approached 60% and only 22 days remain to June 30. Weekly volume ~$18k with No at 97.7¢ shows no meaningful flow challenging the price; smart-money Yes exposure is a trivial $5 stake. Microstructure is clean for a consensus No hold.
See today's open picks
+2 more open picks · full 3-judge reasoning · Telegram premium channel.
Subscribe Now