The Oracle
Ask the Oracle.
Ask any resolvable question about a future event and get back a calibrated probability — not a guess, a number you can hold to account. Every answer is reasoned by an ensemble of eight frontier models, weighed by a supervising model, then published publicly and stamped to the Bitcoin blockchain so it can never be quietly changed after the fact.
$5 per question · every answer public · every answer timestamped
The Oracle wants a sharper question
Suggested wording
The Oracle will answer exactly this question
Once asked, an eight-model ensemble deliberates for ~8 minutes, then the answer is published to the public record and timestamped.
Your question is held
Paid questions open soon. When they do, this one answers for $5 — an eight-model ensemble, deliberated and timestamped. Consider it reserved.
The Oracle declines
The Oracle answers questions about the future. It does not take instructions.
How to ask a good question
The Oracle answers questions that reality can settle. Three things make one resolvable:
- A clear event. One specific thing that either happens or doesn’t — no vibes, no “will AI change everything.”
- Explicit resolution criteria. Say exactly what counts as yes. Name the source that decides it if there’s any doubt.
- A date. Give the question a deadline, so the answer can be scored when the day comes.
Good: “Will SpaceX land humans on the Moon before 1 Jan 2028, per an official NASA or SpaceX announcement?”
Proof of accuracy
The Oracle has a track record you can check. These numbers come straight from the resolved record below — they move honestly as questions settle.
The public record
Every question the Oracle has answered on Polymarket markets — open bets first, then everything that has since resolved. Nothing is ever taken down.
Will the Democratic Party win the ME-01 House seat?
Seven of eight agents converge tightly between 93-96%, with strong consensus on the key evidence: Solid D Cook rating, D+11 PVI, 100% Democratic hold rate since 2008, incumbent running for reelection, weak Republican field. Agent 1 (50%) experienced a complete research failure and produced a meaningless default estimate that should be entirely excluded. Excluding Agent 1, the ensemble mean is...
Will the Democratic Party win the UT-02 House seat?
Seven of eight agents converge strongly on 2-8% probability, with six agents at 2-3%. Agent 7 (50%) had a research failure and should be discarded entirely. The consensus is overwhelming and well-evidenced: under the new court-ordered Utah congressional map (upheld Feb 23, 2026), UT-02 is the successor to Blake Moore's old UT-01 covering northern Utah (Weber/Davis/Cache counties) with an R+38...
Will the Republican Party hold between 200 and 204 House seats after the 2026 midterm elections?
All 8 agents converge tightly on 11-13% (σ=0.008), matching the market price of 11%. The evidence is well-researched and consistent: Republicans hold ~218-220 seats, the 200-204 range requires a 14-18 seat loss, which is plausible but represents just one 5-seat bin in a wide distribution. The Cook floor of 202 seats supports this range as plausible, but Trump's low approval...
Will the Republican Party win the FL-18 House seat?
Seven of eight agents converge strongly on 93-96% probability. Agent 1 (50%) clearly failed its research (0 pages read) and should be discarded entirely. Excluding that outlier, the effective ensemble mean is ~94.6% with very tight agreement. The evidence is overwhelming and consistent across all functioning agents: R+14 PVI, 30-point victory margins, Solid R rating from Cook, 7 consecutive Republican...
Will the Republican Party win the PA-05 House seat?
Seven of eight agents converge strongly on a very low probability (3-8%), with only Agent 1 (Gemini) at 50% due to a research failure (0 pages read, 0 confidence). Excluding this outlier, the remaining 7 agents have a mean of ~4.9% and median of 4%. The evidence is overwhelming and consistent across all functioning agents: Cook PVI D+15, Solid D...
Will the Republican Party win the PA-13 House seat?
Seven of eight agents converge strongly at 93-97% (Agent 1 at 50% is clearly a failed research run with 0 confidence and should be discarded entirely). The remaining seven agents found identical evidence: R+23 PVI, Solid R rating from Cook, Joyce winning 74.2% in 2024, Trump endorsement, no credible Democratic challenger, deeply rural district with no urban centers. This is...
Will the Republican Party win the WV-02 House seat?
All 8 agents converge tightly on 94-99% probability (mean 96.5%, std dev 1.4%), representing near-perfect consensus. The evidence is overwhelming and consistent across all agents: R+30 PVI, 70.8% vote share in 2024, Solid R rating from Cook, no primary challenger, weak Democratic field. The only disagreements are minor calibration differences about how much tail risk to assign to black swan...
Will the Republicans win the New York governor race in 2026?
All 8 agents converge on the same conclusion with remarkable consistency (range 4-8%, std dev 1.3%). Every agent identified the same core evidence: Hochul's commanding 26-point polling lead, record-high favorability, Trump's deep unpopularity in NY, Stefanik's withdrawal leaving a weak GOP nominee, Cook's 'Solid D' rating, and structural Democratic advantages (2:1 registration, no GOP statewide win since 2006). This is...
Will a new country join the Abraham Accords by March 31?
7 of 8 agents cluster between 18-35% (mean ~28.6%), with only Agent 6 at 55% as a clear outlier whose reasoning doesn't justify the deviation. The core evidence is strong and specific: Kazakhstan announced Nov 6, 2025 but hasn't formally signed despite 3+ months and a high-level visit. The Polymarket clarification requires formal signing. The market dropped 11 points in...
Will any xAI Grok model score at least 30% on the FrontierMath Exam?
Agent 3 (8%) misread the resolution date as Feb 28 instead of June 30, creating a misleading outlier. Excluding it, agents cluster 58-82% with mean ~68%. The market at 74.5% is supported by sibling market structure (40% threshold at 72% implies 30% should be higher) and the recent price jump suggesting new information. Grok 4 Heavy already at ~26% means...
Will Germany GDP growth in Q1 2026 be less than or equal to 0.0%?
The well-researched agents (2, 3, 4, 6) who did the GDP level math all converge at 5-6%. Agent 8's higher estimate relies on a factual error about Q1 2025 being strong (it was weak). Agent 7 over-weights historical patterns without accounting for current mathematical constraints. The median of 7.5% is reasonable but the strongest evidence supports ~5-6%. At market price...
Will Luis Fernando Camacho win the 2026 Santa Cruz gubernatorial election?
All 8 agents agree Camacho is the clear frontrunner with confirmed candidacy, incumbency, ruling coalition backing, and a 13-point polling lead. The price spike from 38% to 84.5% is explained by concrete developments (TSE confirmation + legal case dismissal). The main risk is a runoff scenario where he's below 50%, but even in a runoff he'd be favored given his...
Will the Movement for Socialism (MAS-IPSP) win the most gubernatorial elections in Bolivia in 2026?
Seven of eight agents converge strongly on MAS having very low probability (5-18%), with Agent 1 (50%) being an obvious outlier due to research failure (0 pages read). Excluding Agent 1, the mean drops to ~11.4% and median to ~8%. The consensus is overwhelming and well-evidenced: **Where agents agree (high reliability):** - MAS suffered catastrophic collapse in 2025 general elections...
Will Ukraine agree to cede territory to Russia before 2027?
Seven of eight agents cluster between 12-22%, with strong consensus on the key structural barriers. Agent 3 (Grok) is the outlier at 45%, driven by interpreting US pressure and the 28-point plan more aggressively. However, Agent 3's evidence is weaker on the critical resolution criteria distinction. **Where agents agree (high reliability):** - Russia's maximalist demands far exceed what Ukraine could...
Russia x Ukraine ceasefire by end of 2026?
AGENT AGREEMENT: Six agents with substantive research (Agents 3-8) show strong consensus at 30-38%, with four of six at 30-35%. Agents 1 and 2 produced no research and defaulted to 50% - these should be entirely discarded. The effective ensemble mean excluding failed agents is ~32.5%, with tight clustering. KEY CONSENSUS FINDINGS: (1) Russia explicitly refuses ceasefire before comprehensive political...
Resolved
Khamenei out as Supreme Leader of Iran in 2026?
All 8 agents unanimously agree at 99% probability. The event has already occurred - Khamenei was killed on February 28, 2026 in confirmed US-Israeli airstrikes. This is verified by Iranian state media, multiple independent international news organizations, US and Israeli officials, and a successor has already been appointed. The market has been at 100% for at least 7 consecutive days....
Tesla launches unsupervised full self driving (FSD) by March 31?
7 of 8 agents converge at 99%, with only Agent 8 (grok-4) at 85% due to weighing the 'not truly unsupervised' criticism more heavily. However, Agent 8 appears to have missed or underweighted the critical fact that the market has ALREADY RESOLVED YES via UMA oracle on January 29, 2026, with the dispute window closed over 7 weeks ago. This...
U.S. x Iran Military Engagement by June 30, 2026?
7 of 8 agents converge at 99% probability. Agent 8 (Grok) returned 50% due to research failure (0 confidence, incomplete search) and should be entirely discounted - it's a non-informative default. The ensemble mean of 92.88% is artificially dragged down by this single failed agent; the true informed consensus is 99%. The evidence is overwhelming and independently verified across agents:...
Khamenei out as Supreme Leader of Iran in 2026?
7 out of 8 agents independently found overwhelming, mutually corroborating evidence that Khamenei was killed on February 28, 2026 in US-Israeli airstrikes. The evidence includes confirmation from Iranian state media itself, all major international news organizations, a declared 40-day mourning period, and the appointment of a successor (Mojtaba Khamenei). Agent 1 (GPT-5.2) returned 50% due to a research failure (0...
Tesla launches unsupervised full self driving (FSD) by March 31?
7 out of 8 agents (all except Agent 1, which failed to complete research) converge on 98-99% probability. The key finding is unanimous among functioning agents: this market has ALREADY RESOLVED YES via UMA oracle on January 29, 2026, based on Tesla's January 22 announcement of unsupervised robotaxi rides in Austin. The dispute window closed weeks ago. The market has...
U.S. x Iran Military Engagement by June 30, 2026?
7 of 8 agents independently arrived at 99% probability after conducting separate research. Agent 1 (50%) had a research failure with 0 searches and 0 pages read, making it a complete outlier that should be disregarded. The remaining 7 agents all found the same overwhelming evidence: Operation Epic Fury commenced February 28, 2026, involving thousands of U.S. kinetic strikes directly...
Tesla launches unsupervised full self driving (FSD) by March 31?
7 of 8 agents converge on 99% probability. The market has already resolved YES via the UMA oracle on January 29, 2026, based on Tesla's January 22, 2026 launch of unsupervised robotaxi rides in Austin. The dispute window has closed and the resolution is final. The market has been at 100% for 7+ weeks. Agent 6 (Grok) is the sole...
U.S. x Iran Military Engagement by June 30, 2026?
7 of 8 agents independently found overwhelming evidence of an active US-Iran war beginning February 28, 2026 (Agent 1 failed to complete research). The evidence is extraordinarily strong - confirmed by CENTCOM, White House, Reuters, BBC, Wikipedia, Britannica, ISW, and numerous other sources. Operation Epic Fury involved ~900 strikes in the first 12 hours, 7,800+ total strikes, Iranian retaliation against...
Khamenei out as Supreme Leader of Iran in 2026?
7 out of 8 agents independently arrived at 99% probability, with the 8th agent (Agent 8) producing an incomplete/failed analysis at 50% with 0 confidence - this is clearly a data artifact and should be discarded. The remaining 7 agents found identical evidence: Khamenei was killed Feb 28, 2026 in US-Israeli strikes, confirmed by Iranian state media, multiple international outlets,...
Tesla launches unsupervised full self driving (FSD) by March 31?
7 of 8 agents (excluding the failed Agent 8 and the outlier Agent 7) converge on 97-99% probability. The critical finding is that this market has ALREADY RESOLVED YES via UMA oracle on January 29, 2026, based on Tesla's January 22, 2026 announcement of unsupervised robotaxi rides in Austin. The dispute window has closed, making the resolution final. The market...
U.S. x Iran Military Engagement by June 30, 2026?
Seven of eight agents independently arrived at 99% probability with high confidence. Agent 8 (50%, 0 confidence) had a research failure and should be discarded entirely. The consensus is overwhelming: the U.S. launched kinetic strikes on Iranian territory starting February 28, 2026, Iran retaliated, and the conflict is ongoing. This clearly meets the resolution criteria ('any U.S. military kinetic strike...
Khamenei out as Supreme Leader of Iran in 2026?
All 8 agents unanimously agree at 99% probability. The event has already occurred: Khamenei was killed on February 28, 2026 in US-Israeli strikes, confirmed by Iranian state media (which declared 40 days of mourning), multiple top-tier international outlets (Reuters, BBC, CNN, Al Jazeera, NPR, Forbes, CNBC), and government officials (Trump, Netanyahu). A successor (Mojtaba Khamenei) has already been named and...
Tesla launches unsupervised full self driving (FSD) by March 31?
7 out of 8 agents (all except Agent 4) converge on 97-99% probability, recognizing that this market has ALREADY RESOLVED YES via UMA oracle on January 29, 2026. The resolution was based on Tesla's January 22, 2026 announcement of unsupervised robotaxi rides in Austin. The market has been at 100% for 7+ weeks with no active dispute. Agent 4 (3%)...
U.S. x Iran Military Engagement by June 30, 2026?
All 8 agents independently arrived at exactly 99% probability with high confidence (0.93-0.99). Every agent found the same overwhelming evidence: Operation Epic Fury began February 28, 2026 with joint US-Israeli airstrikes on Iran, over 8,000 US military targets struck, Iranian retaliatory missile attacks on US bases, confirmed US casualties, and ongoing escalation as of March 22, 2026. This is not...
Khamenei out as Supreme Leader of Iran in 2026?
All 8 agents unanimously agree at 99% probability. The event has already occurred - Khamenei was assassinated on February 28, 2026, confirmed by every major international news outlet and Iranian state media itself. A successor has been appointed. This is not a forecast but a confirmation of a past event. The market is correctly priced at 100%. However, despite overwhelming...
Tesla launches unsupervised full self driving (FSD) by March 31?
All 8 agents agree this market has already resolved YES via UMA oracle on ~January 29, 2026, based on Tesla's January 22, 2026 launch of unsupervised robotaxi rides in Austin. The range of estimates (97-99%) reflects only minor disagreement about the infinitesimal probability of an unprecedented UMA oracle reversal. Agent 8 provided the most balanced evidence noting chase cars and...
U.S. x Iran Military Engagement by June 30, 2026?
All 8 agents unanimously estimate 99% probability. The evidence is overwhelming: a full-scale U.S.-Iran war began on February 28, 2026, with U.S. kinetic strikes on Iranian territory and Iranian retaliatory missile strikes on U.S. bases across multiple countries. This clearly meets the market's definition of 'military engagement involving the use of force such as missile strikes, artillery fire, exchange of...
Khamenei out as Supreme Leader of Iran in 2026?
All 8 agents unanimously agree at 99% probability. The event has already occurred: Khamenei was killed on February 28, 2026 in confirmed US-Israeli airstrikes. This was verified by Iranian state media, every major international news outlet, the White House, and a successor has already been appointed. The market is correctly priced at 100%. There is zero actionable edge - the...
Tesla launches unsupervised full self driving (FSD) by March 31?
All 8 agents agree this market has already resolved YES via UMA oracle around January 29, 2026, following Tesla's January 22, 2026 unsupervised robotaxi launch in Austin. The market has been at 100% for over 7 weeks. The only disagreement is minor - Agent 7 at 90% appears to have not fully weighted the fact that the market is already...
U.S. x Iran Military Engagement by June 30, 2026?
All 8 agents unanimously estimate 99% probability. The evidence is overwhelming and independently verified across dozens of credible sources: the U.S. and Israel launched Operation Epic Fury on February 28, 2026, striking over 2,000 targets in Iran, killing Supreme Leader Khamenei, and Iran retaliated with missile attacks on U.S. bases in 6 countries. The conflict is ongoing on Day 21...
Khamenei out as Supreme Leader of Iran in 2026?
All 8 agents unanimously agree at 99% probability. The event has already occurred: Khamenei was killed on February 28, 2026, confirmed by every major international news organization and Iranian state media itself. A successor has already been appointed. The market is correctly priced at 100%. There is zero disagreement among agents on the core facts. The only reason not to...
Tesla launches unsupervised full self driving (FSD) by March 31?
All 8 agents agree on the core facts: Tesla launched unsupervised robotaxi rides in Austin on January 22, 2026, the UMA oracle resolved the market YES around January 29, 2026, and the market has been at 100% for 7+ weeks. The slight variation in estimates (95-99%) comes from Agent 7 giving more weight to the substantive controversy about whether Tesla's...
U.S. x Iran Military Engagement by June 30, 2026?
All 8 agents unanimously estimate 99% probability. The evidence is overwhelming and independently verified across dozens of authoritative sources: the 2026 Iran War began on February 28, 2026 with coordinated US-Israeli airstrikes on Iran, Iran has retaliated against US military bases, US service members have been killed, and the conflict is ongoing as of March 19, 2026. This clearly and...
Israel strike on Yemen by March 31, 2026?
After excluding Agent 1 (who misunderstands resolution criteria by counting 2025 strikes), the remaining 7 agents cluster around 5-72% with most in the 17-52% range. The adjusted mean is ~32%. The core question is whether Houthis break 3+ weeks of restraint in 13 days. Strong evidence of deterrence (leadership decapitation fear, logistical exhaustion, Iran holding them in reserve) vs. genuine...
Khamenei out as Supreme Leader of Iran in 2026?
All 8 agents unanimously assign 99% probability. The event has already occurred: Khamenei was killed on February 28, 2026 in joint US-Israeli airstrikes, confirmed by Iranian state TV, BBC, Reuters, CNN, Al Jazeera, and numerous other credible international outlets. A successor (Mojtaba Khamenei) has already been appointed by the Assembly of Experts. This clearly satisfies the resolution criteria of being...
Tesla launches unsupervised full self driving (FSD) by March 31?
This market has already resolved YES via UMA oracle around January 29, 2026. The critical question is NOT whether Tesla truly launched unsupervised FSD, but whether the already-finalized market resolution will be reversed. 6 of 8 agents (Agents 1,2,3,5,6,7) correctly identified this as an already-resolved market and gave probabilities of 97-99%. Agents 4 and 8 appear to have analyzed the...
Khamenei out as Supreme Leader of Iran in 2026?
All 8 agents unanimously agree at 99% probability. The event has already occurred - Ali Khamenei was killed on February 28, 2026 in US-Israeli airstrikes, confirmed by every major international news outlet and Iranian state media itself. A formal succession process was completed with Mojtaba Khamenei elected as the new Supreme Leader. This is not a forecast but a confirmation...
Tesla launches unsupervised full self driving (FSD) by March 31?
This market has already resolved to YES. The split among agents is not about what will happen but about whether the resolution was 'correct' in substance. Agents 1, 2, 4, 5, 6 (5 of 8) correctly identify that the market has already resolved YES via UMA oracle around January 29, 2026, and assign 97-99% probability. Agents 3, 7, 8 argue...
Tesla launches unsupervised full self driving (FSD) by March 31?
7 of 8 agents agree at 97-99% that this market has already resolved YES. Agent 8 (2%) appears to be evaluating whether Tesla truly launched unsupervised FSD rather than whether the market resolves YES — a category error. The market resolved via UMA oracle on Jan 29, 2026 based on Tesla's Jan 22 Austin robotaxi announcement. Despite legitimate controversy (chase...
Tesla launches unsupervised full self driving (FSD) by March 31?
6 of 8 informed agents agree the market has already resolved YES. Agent 1 had a research failure (0 searches) and should be discarded. Agent 7 acknowledged their estimate would be much higher if the market already resolved, which it has. The ensemble mean of 80.25% is dragged down by these two outlier agents and is not representative. The true...
Tesla launches unsupervised full self driving (FSD) by March 31?
All 8 agents agree this market has effectively already resolved YES, with probabilities ranging from 85-99%. The core facts are undisputed: Tesla launched robotaxi rides in Austin on January 22, 2026 without an in-car safety monitor, the UMA oracle resolved the market YES around January 29, and the price has been at 100% since. The only dissent comes from Agent...
US x Iran nuclear talks resume by March 31?
All 8 agents unanimously assign 99% probability. The event has already occurred: US-Iran nuclear talks resumed on February 6, 2026 in Oman, with a second round February 17-19 in Geneva and a third round scheduled for February 26-27. Official representatives (Iranian FM Araghchi, US envoy Witkoff) participated in nuclear-focused discussions. Multiple independent credible sources confirm this. The market is priced...
1H Spread: Pistons (-5.5)
All 8 agents unanimously agree at 99% probability. The game was played on November 24, 2025, with the Pistons leading 71-55 at halftime — a +16 margin that covers the -5.5 spread by 10.5 points. Multiple independent sources (ESPN box score gameId 401810138, Basketball Reference 202511240IND, Reuters, MLive) confirm the halftime score. Sibling markets are fully consistent: 1H Moneyline at...
OG Anunoby: Points O/U 17.5
7 of 8 agents independently confirmed via multiple authoritative sources (ESPN, NBA.com, CBS Sports, Basketball-Reference, StatMuse, Sportsnet) that the game was played on February 4, 2026, and OG Anunoby scored 20 points. The detailed stat line (7-17 FG, 4-9 3PT, 2-2 FT, 47-48 minutes in a double-overtime game) is consistent across all sources. Agent 5 (gpt-5.2) was the sole outlier...
Spread: North Texas (-26.5)
All 8 agents unanimously agree at 99% probability. The game was played on October 24, 2025, and North Texas won 54-20, a 34-point margin that covers the -26.5 spread by 7.5 points. This is confirmed by multiple independent authoritative sources (ESPN, CBS Sports, Fox Sports, NCAA.com, official team and conference sites). The market is already resolved at 100%. There is...