AIs disagree on chip index for the week

What you can do here

Make your own forecast before the AIs →
See which AI is leading →
AIs against the crowd (Polymarket, Kalshi) →
Try a portfolio of your own size →
Live market prices →
The league and the Pro league →
What the AIs would do in each market →
How we count →
For pros →
Open data, RSS and webhook →
Score of the day and the archive →
The ratings of the day on Top and their archive →

MoneyAI Portfolio League: the leaders^‡ — virtual portfolios, $100,000 each at the start

ChatGPT $98,740 −1.26%
MiMo (Xiaomi) $98,317 −1.68%
StepFun $98,018 −1.98%

S&P 500 over the same days: −1.12% The whole league → Check these accounts → · part 2 every trade with its price and cost; the accounts day by day (CSV)

As of Nov 11 · sessions so far: 18 · the AIs and the S&P 500 after costs

MoneyThe Council's virtual portfolio since Oct 19 (18 sessions, after the league's costs): $100,000 → $97,069; the S&P 500 bought and held → $98,885 (from the 10-19 open, first price 09:30–09:45 ET, after costs).^‡ Pick the day →

DirectionThe result for Wed, Nov 11, 2026: the AIs called the direction of 7 of 10 markets — S&P 500, Nasdaq-100, chip index, dollar index, fear index (VIX), WTI crude oil, Brent crude oil. Score of the day → · Check this score → the day’s rows: each AI’s call, the reference price, the close and the move (CSV)

The score in detail

Last score · Wed, Nov 11, 2026: the AIs' called 7 of 10 indicators right. ClosenessClosest most often that day, level: ChatGPT, DeepSeek, Kimi, Meta AI, MiMo (Xiaomi), Nemotron, Amazon Nova, Qwen, StepFun, Z.ai — 1 of 10 indicators each. Score of the day →

History and accuracy (Brier)

sessions scored: 18 · the consensus' 0.243 (95% interval 0.232–0.255) (approximate)

What was counted: whether each indicator ended the session above or below its reference price, against the AIs' consensus call. Where the consensus missed: gold, copper, Bitcoin. Calling the direction and making virtual money are two separate scores: in the portfolio league, AI portfolios ahead of the S&P 500: 0 · the Council $97,069 against the S&P 500 $98,885.

How reliable are the forecasts ▾

The consensus Brier by horizon (a coin is 0.25; in brackets the single models’ mean): Day 0.243 (0.246) · Week 0.252 (0.257)
Brier of the Council’s consensus (day): 0.243 · 95% interval by forecast 0.236–0.250 · by whole sessions: 0.232–0.255
Oct 19, 2026 – Nov 11, 2026: the Council’s Brier skill against the crowd is −0.28 (95% interval −0.43 to −0.12), 18 scored sessions. Brier skill is the share by which the Council’s error is smaller than the crowd’s: above zero, the Council is more accurate than the crowd; below zero, less accurate; the higher, the better.
They said 57% — it came true 55% (2186 day forecasts over 18 sessions, since Oct 19).
effective voices, estimate by correlation (scored issues, 1 day): 1.74 · 1.46 · 2.23 · 1.60 · 2.57 · 1.93 · 3.00 · 2.61 · 3.32 · 1.67 · 1.82 · 4.05 · 1.49 · 1.73 · 1.68 · 1.63 · 3.47 · 3.04 — mean 2.28
citations of a site another AI also cites: 46%, of the same link: 6% · cited most: reuters.com 21%
This number shows how similar the models’ probabilities are, not whether their errors are independent · by the correlation of errors: 2.93
Models can rely on the same news. Agreement is not independent confirmation; the heat shows how far their actions differ. Effective voices are an estimate from how the probabilities move together, including the link between the markets themselves (the S&P 500, the Nasdaq and the Dow move together); they are not proof that the models are independent.
The rule of the game: to tell 55% right calls from a coin takes about 209 sessions ≈ 10 months; the ranking comes earlier and sharpens every day. (calculated assuming the forecasts of one day are linked, ρ≈0.2; from 5 sessions — with the link measured in the book)
Check it in three steps: the question word for word, the day’s score, every forecast in CSV: 1 · 2 · 3
Pairs of models: the Brier difference on the same events with a 95% interval by sessions; one model is better than another only when the interval leaves out 0 — such pairs now: 2 of 91 · council-pairs.csv

Track record: 30 / 90 / 180 days ▾

The windows of 30, 90 and 180 days open at 30 scored sessions: now 18 of 30.

Files: council-stats.json (track_record)

Calls are closed. The result: Fri, Nov 13, 2026, 5:00 PM . Score of the day: the AIs' consensus called 8 of 10 indicators right.

Your call before the AIs: where will chip index be by Nov 13, 2026 (week)?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (9,474.17) to the value at Fri, Nov 13, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the threshold is a draw.

Get the morning issue and the score in the free Poly app →

Ask the Council about any ticker — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

↑ Top

Who would you trust with a virtual $100,000? →

Archive: the issue of Nov 7, 2026 · v1 10:48 AM ET Latest issue →

🔒 Every forecast is sealed before it counts · Verify the signature · CSV ·

hash and time

Every forecast is before it counts: signed by RFC 3161 timestamp services (DigiCert, FreeTSA, Sectigo) before the open; the hash 1b631feccb39… is then anchored in Bitcoin (OpenTimestamps)

The Council's virtual portfolio from the day you pick^‡

Virtual $100,000 put into the Council's portfolio at the close before the chosen day, against the same sum in the S&P 500 bought and held; the league's own costs are inside.

From the day:

By Nov 11: the Council $97,069 · the S&P 500 $98,885 (from the 10-19 open, first price 09:30–09:45 ET, after costs)

As of Nov 11 · sessions so far: 18

Best start day so far: Nov 11 → the Council $100,096 · the S&P 500 $101,319 (from the 11-10 close, the Council and the S&P 500 alike, without the entry cost)

Weakest start day so far: Oct 22 → the Council $96,715 · the S&P 500 $98,282 (from the 10-21 close, the Council and the S&P 500 alike, without the entry cost)

JSON · CSV

A public league of AI forecasts with a history no one can rewrite.

Words on this page

b.p. — a basis point, 0.01 percentage point (a yield from 4.10% to 4.15% is +5 b.p.)
VIX — the market's expected swings over 30 days
US 10-year and VIX: the AIs bet on the indicator itself (through a fund), not on bonds or stocks. “Keep the bet” — keep the bet as it is (not “hold” a security); “Toward a rise, on a dip” — toward a rise, entered only if it dips to the model's level; “Strongly toward a rise” — stronger toward a rise; “Toward a fall” — toward a fall; “Strongly toward a fall” — stronger toward a fall.
“buy on dips” — add only if the price falls to the model's own level
“trim” — sell part, keep the rest
ET — New York time (the exchange's clock)
Brier — how far the stated odds were from what happened: 0 perfect, 0.25 a coin
consensus — what most of the AIs say, weighted so one family cannot outvote the others
horizon — the day, the week, the month, six months
a draw — a move too small to call (under a tenth of a typical day)

About this issue: times and rules

Versions of the issue, 11-07: v1 10:48 AM ET · answers taken until Mon, Nov 9, 2026, 9:30 AM ET · the same for everyone at each version

Next issue: Sun, Nov 8, 2026, 5:15 PM ET — the Sunday issue: the week’s radar

Weekend issue: the week, the month and 6 months count from Monday's open; Bitcoin counts from the seal.

The AIs’ answers given by Mon, Nov 9, 2026, 9:30 AM ET count from the opening price; an answer given after that counts from the price at the moment it was given. Your own call is taken until Mon, Nov 9, 2026, 9:30 AM ET.

Before the New York open, the world’s leading AIs say what they would do with each of 12 key markets if it were their own virtual money. Every forecast is sealed before it starts counting, so no one can rewrite it later. After the close, the score shows who was right.

Now

Each tile: the price and its change from the previous session’s close.

Season 1, session 18 of 126

S&P 5007,619.38▲ +1.31%close Wed, Nov 11, 2026 · last quote 4:42 PM ET Nasdaq-10028,374.25▲ +2.28%close Wed, Nov 11, 2026 · last quote 5:15 PM ET Dow Jones52,486.49▲ +0.49%close Wed, Nov 11, 2026 · last quote 4:43 PM ET Chip indexDispute of the day10,564.11▲ +5.66%close Wed, Nov 11, 2026 · last quote 5:15 PM ET US 10-year4.998▲ +1.5 bp · yield ↑ · bonds get cheaperclose Wed, Nov 11, 2026 · last quote 2:59 PM ET Dollar index102.92▲ +0.014%7:54 AM ET · delay 15 min Fear index (VIX)18.35▲ +4.91%7:49 AM ET · delay 15 min Gold3,941.10▼ −1.49%7:54 AM ET · delay 15 min WTI crude oil85.78▼ −0.87%7:54 AM ET · delay 15 min Brent crude oil92.11▼ −0.70%7:54 AM ET · delay 15 min Copper6.202▼ −1.39%7:54 AM ET · delay 15 min Bitcoin72,794▼ −1.27%8:04 AM ET · delay 0 min

Futures (Thu, 7:54 AM ): S&P 500 (ES) ▼ −0.43% · Nasdaq-100 (NQ) ▼ −0.98% · Dow (YM) ▲ +0.080% · Russell 2000 (RTY) ▼ −0.15%

What the markets expect: Fed, Dec 9, 2026: +0.25 76%, no change 22% · Fed, Jan 27, 2027: no change 74%, +0.25 18%, −0.25 4% · inflation, next release ≈ 3.7%

Coming up: Thu, Nov 12, 2026 8:30 AM ET — weekly jobless claims (US) · Fri, Dec 4, 2026 8:30 AM ET — the US jobs report · Wed, Dec 9, 2026 2:00 PM ET — the Fed's decision (the 10-year yield, the dollar, the indexes, gold) · Thu, Dec 10, 2026 8:30 AM ET — US inflation, CPI (the yield, the dollar, gold, the indexes) · moves first: 10-year US Treasury yield, dollar index, S&P 500, gold

The AIs against the crowd

A close question people bet real money on at and : the AIs’ own odds of “Up” and the crowd’s price (each market’s day is in the card). A daily crowd question exists for 3 of the 12 markets. A crowd question counts from its own reference — on Kalshi, above the previous close at 4:00 PM; the score of the day counts from the open.

The same question as the crowd (the very market, its rules and its end):

S&P 500: AIs: 54% chance of a rise · crowd 54% · effective voices ≈ 2.3 (mean of the issues) · thin market

For orientation — not quite the same question (another window, or Kalshi's price read between the two nearest strikes):

WTI crude oil: AIs: 51% chance of a rise · crowd 72% · effective voices ≈ 2.3 (mean of the issues) · another window: this Polymarket question ends Mon 6:00 PM , not at our event's end · thin market
Nasdaq-100: AIs: 63% chance of a rise · crowd 52% · effective voices ≈ 2.3 (mean of the issues) · thin market

In one line: the AIs argue moderately, are neutral overall, their confidence is low.

0 — all agree, 100 — an even split
citations of a site another AI also cites: 46%, of the same link: 6% · cited most: reuters.com 21%

Models can rely on the same news. Agreement is not independent confirmation; the heat shows how far their actions differ.

Direction

51%

neutral · the AIs' average odds of a rise today over the 12 markets · effective voices ≈ 2.3 (mean of the issues)

Confidence

How far the AIs' day odds are from 50/50, on average: 0 — a coin, 100 — all certain.
Each model on its own, by : Day 9 · Week 13 · Month 15 · 6 months 22.
They said 57% — it came true 55% (2186 day forecasts over 18 sessions, since Oct 19, 2026).

Dispute of the day

Chip index, week: 43% of US AIs say up, 29% of Chinese AIs

Chip index Nasdaq-100 WTI crude oil 10-year US Treasury yield

Since the last issue: S&P 500 — the turned to hold; Nasdaq-100 — the consensus turned to hold; Chip index — the consensus turned to hold; 10-year US Treasury yield — the consensus turned to a rise. Changed side on the day: Nemotron + Mercury (search) (Nasdaq-100, chip index, fear index (VIX)); Amazon Nova (gold, copper); Meta AI (fear index (VIX), WTI crude oil, Brent crude oil); Gemini (Nasdaq-100); Z.ai (10-year US Treasury yield). New split between the US and Chinese AIs: gold, WTI crude oil, Brent crude oil, copper, Bitcoin.

Past forecasts scored since the last issue: Fri, Nov 6, 2026 — the consensus was right on 2 of 8.

Chips next week split the room: Claude, ERNIE (Wenxin), Amazon Nova and Meta AI would buy a dip, while ChatGPT, Gemini, Grok, DeepSeek, Kimi and StepFun would trim.
Nemotron and Z.ai would sell chips outright. Qwen would buy them hard.
The most surprising call is Qwen. It wants to buy hard on the Nasdaq-100, chips, gold, oil and copper for the week, while Z.ai wants to sell the Nasdaq-100 and chips.
On the VIX fear gauge for the week, Nemotron bets it jumps. Qwen, ERNIE (Wenxin), Amazon Nova and Z.ai bet it falls. On the 10-year yield, ERNIE (Wenxin) bets it drops, while Nemotron, Amazon Nova and Z.ai bet it climbs.
They agree on the calm: nearly all just hold the Dow for the day, and most hold the S&P 500, the dollar and gold for the day. Most would buy dips in the S&P 500 and copper over the week.

What the Council would do

Forecasts of the leading AIs of the USA and China for 12 markets over four horizons. Where a model has no new answer, its position is unchanged.

Indicator	Now	Day	Week	Month	6 months	AIs vs the crowd
S&P 500 · other and the crowd ▾	7,619.38▲ +1.31%	Hold · 71% of the AIs chose itaverage chance of a rise: USA 54% · China 54%(effective voices ≈ 2.3 (mean of the issues))	Week Buy on dips · 64% of the AIs chose itaverage chance of a rise: USA 54% · China 52%(effective voices ≈ 2.8 (mean of the issues))	Month Hold · 43% of the AIs chose itaverage chance of a rise: USA 53% · China 48%(effective voices ≈ 3.2 (mean of the issues))	6 months Buy on dips · 50% of the AIs chose itaverage chance of a rise: USA 63% · China 58%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd AIs: 54% chance of a rise · crowd 54% · effective voices ≈ 2.3 (mean of the issues) ↗ price at Sat 9:12 AM , $59 traded thin market
Nasdaq-100 · other horizons and the crowd ▾	28,374.25▲ +2.28%	Hold · 57% of the AIs chose itaverage chance of a rise: USA 50% · China 55%(effective voices ≈ 2.3 (mean of the issues))	Week Trim · 43% of the AIs chose itaverage chance of a rise: USA 47% · China 46%(effective voices ≈ 2.8 (mean of the issues))	Month Buy on dips · 43% of the AIs chose itaverage chance of a rise: USA 47% · China 45%(effective voices ≈ 3.2 (mean of the issues))	6 months Buy on dips · 57% of the AIs chose itaverage chance of a rise: USA 62% · China 56%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd AIs: 63% chance of a rise · crowd 52% · effective voices ≈ 2.3 (mean of the issues) ↗ price at Sat 9:12 AM ET, $683 traded thin market
Dow Jones · other horizons and the crowd ▾	52,486.49▲ +0.49%	Hold · 86% of the AIs chose itaverage chance of a rise: USA 52% · China 50%(effective voices ≈ 2.3 (mean of the issues))	Week Hold · 57% of the AIs chose itaverage chance of a rise: USA 55% · China 53%(effective voices ≈ 2.8 (mean of the issues))	Month Hold · 43% of the AIs chose itaverage chance of a rise: USA 57% · China 55%(effective voices ≈ 3.2 (mean of the issues))	6 months Buy on dips · 50% of the AIs chose itaverage chance of a rise: USA 63% · China 58%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd no price read
Chip index · other horizons and the crowd ▾	10,564.11▲ +5.66%	Hold · 57% of the AIs chose itaverage chance of a rise: USA 52% · China 50%(effective voices ≈ 2.3 (mean of the issues))	Week Trim · 43% of the AIs chose itaverage chance of a rise: USA 47% · China 44%(effective voices ≈ 2.8 (mean of the issues))	Month Hold · 43% of the AIs chose itaverage chance of a rise: USA 45% · China 43%(effective voices ≈ 3.2 (mean of the issues))	6 months Buy on dips · 79% of the AIs chose itaverage chance of a rise: USA 62% · China 58%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd no market
10-year US Treasury yield · other horizons and the crowd ▾	4.998▲ +1.5 bp · yield ↑ · bonds get cheaper	Toward a rise, on a dip · 57% of the AIs chose itaverage chance of a rise: USA 58% · China 56%(effective voices ≈ 2.3 (mean of the issues))	Week Toward a rise, on a dip · 50% of the AIs chose itaverage chance of a rise: USA 60% · China 55%(effective voices ≈ 2.8 (mean of the issues))	Month Toward a rise, on a dip · 29% of the AIs chose itaverage chance of a rise: USA 52% · China 61%(effective voices ≈ 3.2 (mean of the issues))	6 months Toward a fall · 50% of the AIs chose itaverage chance of a rise: USA 39% · China 58%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd no market
Dollar index · other horizons and the crowd ▾	102.92▲ +0.014%	Hold · 79% of the AIs chose itaverage chance of a rise: USA 52% · China 53%(effective voices ≈ 2.3 (mean of the issues))	Week Hold · 36% of the AIs chose itaverage chance of a rise: USA 52% · China 56%(effective voices ≈ 2.8 (mean of the issues))	Month Hold · 64% of the AIs chose itaverage chance of a rise: USA 49% · China 55%(effective voices ≈ 3.2 (mean of the issues))	6 months Trim · 64% of the AIs chose itaverage chance of a rise: USA 41% · China 45%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd no price read
Fear index (VIX) · other horizons and the crowd ▾	18.35▲ +4.91%	Toward a fall · 50% of the AIs chose itaverage chance of a rise: USA 44% · China 42%(effective voices ≈ 2.3 (mean of the issues))	Week Toward a fall · 50% of the AIs chose itaverage chance of a rise: USA 38% · China 42%(effective voices ≈ 2.8 (mean of the issues))	Month Strongly toward a fall · 43% of the AIs chose itaverage chance of a rise: USA 35% · China 40%(effective voices ≈ 3.2 (mean of the issues))	6 months Strongly toward a fall · 50% of the AIs chose itaverage chance of a rise: USA 29% · China 39%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd no market
Gold · other horizons and the crowd ▾	3,941.10▼ −1.49%	Hold · 79% of the AIs chose itaverage chance of a rise: USA 52% · China 48%(effective voices ≈ 2.3 (mean of the issues))	Week Buy on dips · 57% of the AIs chose itaverage chance of a rise: USA 55% · China 51%(effective voices ≈ 2.8 (mean of the issues))	Month Buy on dips · 57% of the AIs chose itaverage chance of a rise: USA 59% · China 54%(effective voices ≈ 3.2 (mean of the issues))	6 months Buy more · 50% of the AIs chose itaverage chance of a rise: USA 66% · China 62%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd no price read
WTI crude oil · other horizons and the crowd ▾	85.78▼ −0.87%	Hold · 57% of the AIs chose itaverage chance of a rise: USA 53% · China 49%(effective voices ≈ 2.3 (mean of the issues))	Week Buy on dips · 43% of the AIs chose itaverage chance of a rise: USA 53% · China 50%(effective voices ≈ 2.8 (mean of the issues))	Month Hold · 57% of the AIs chose itaverage chance of a rise: USA 45% · China 50%(effective voices ≈ 3.2 (mean of the issues))	6 months Hold · 29% of the AIs chose itaverage chance of a rise: USA 44% · China 55%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd AIs: 51% chance of a rise · crowd 72% · effective voices ≈ 2.3 (mean of the issues) ↗ price at Sat 9:12 AM ET, $42 traded another window: this question ends Mon 6:00 PM ET, not at our event's end thin market
Brent crude oil · other horizons and the crowd ▾	92.11▼ −0.70%	Hold · 57% of the AIs chose itaverage chance of a rise: USA 53% · China 49%(effective voices ≈ 2.3 (mean of the issues))	Week Hold · 36% of the AIs chose itaverage chance of a rise: USA 52% · China 50%(effective voices ≈ 2.8 (mean of the issues))	Month Hold · 50% of the AIs chose itaverage chance of a rise: USA 46% · China 50%(effective voices ≈ 3.2 (mean of the issues))	6 months Hold · 29% of the AIs chose itaverage chance of a rise: USA 45% · China 54%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd no market
Copper · other horizons and the crowd ▾	6.202▼ −1.39%	Hold · 71% of the AIs chose itaverage chance of a rise: USA 53% · China 49%(effective voices ≈ 2.3 (mean of the issues))	Week Buy on dips · 64% of the AIs chose itaverage chance of a rise: USA 56% · China 53%(effective voices ≈ 2.8 (mean of the issues))	Month Buy on dips · 50% of the AIs chose itaverage chance of a rise: USA 59% · China 52%(effective voices ≈ 3.2 (mean of the issues))	6 months Buy on dips · 50% of the AIs chose itaverage chance of a rise: USA 64% · China 60%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd no market
Bitcoin · other horizons and the crowd ▾	72,794▼ −1.27%	Hold · 57% of the AIs chose itaverage chance of a rise: USA 48% · China 51%(effective voices ≈ 2.3 (mean of the issues))	Week Hold · 50% of the AIs chose itaverage chance of a rise: USA 45% · China 49%(effective voices ≈ 2.8 (mean of the issues))	Month Buy on dips · 43% of the AIs chose itaverage chance of a rise: USA 53% · China 48%(effective voices ≈ 3.2 (mean of the issues))	6 months Buy on dips · 64% of the AIs chose itaverage chance of a rise: USA 63% · China 59%(effective voices ≈ 1.6 (mean of the issues) · on deviations from climatology ≈ 1.22)	AIs vs the crowd no price read

Heat map: market × model

Colour is the side (green — up, red — down), strength is the confidence; ▲ up, ▼ down, ● even. Pick the horizon.

Day Week Month 6 months

Show: (points; the colour stays the side)

Tap a cell of the map: you get a link to that slice — market × model × horizon × this day.

Day

	ChatGPT	Claude	Gemini	Grok	Meta AI	Nemotron	Amazon Nova	DeepSeek	Qwen	Kimi	Z.ai	ERNIE	StepFun	MiMo
S&P 500	▲55	▲54	▲58	▲54	▼47	▲55	●51	●52	▲55	▲58	▲56	▲54	●52	▲55
Nasdaq-100	▲56	▲54	▼44	●52	▼46	▼42	●52	●52	▲55	▲57	▲55	▲55	▲53	▲58
Dow Jones	▲53	▲53	▲59	●51	●52	●52	●50	▼47	▲60	●50	●50	●51	●50	▼47
Chip index	▲58	▲54	▼41	▲55	▲55	▼38	▲55	●51	▼45	▼47	●52	▲62	●48	▲53
10-year US Treasury yield	▲55	●50	▲58	▲58	▲58	▲60	▲60	●50	▲60	▲60	▲57	▼42	▲55	▲57
Dollar index	▲53	●50	▲59	●52	●52	▼45	●52	●52	●50	▲56	●52	▲53	●52	▲54
Fear index (VIX)	▼42	▼43	▼40	▼42	▲55	▲58	▼30	▼44	▼40	●48	▼38	▼28	▼40	▼46
Gold	▲54	●51	▲57	●48	●48	●51	▲55	●51	▼45	▼47	●50	●48	●50	▼44
WTI crude oil	▲54	▲56	▲57	●51	●52	▲54	●51	▼45	●50	▼47	▲55	●52	▼46	●52
Brent crude oil	▲55	▲55	▲56	●50	●52	▲53	●50	▼45	●50	▼47	▲54	▲53	▼46	●51
Copper	▲55	●52	●52	▲56	●49	●51	▲55	●48	▼45	▲56	●52	▼47	●48	●48
Bitcoin	▼45†	●52†	▼43†	●51†	▼47†	▼45†	▲60†	●48†	▲55†	▲56†	▼47†	▲55†	●50†	▼45†

Week

	ChatGPT	Claude	Gemini	Grok	Meta AI	Nemotron	Amazon Nova	DeepSeek	Qwen	Kimi	Z.ai	ERNIE	StepFun	MiMo
S&P 500	▼47	▲55	▲55	▼46	▲54	▲58	▲60	▼46	●50	▲55	▲55	●52	▲55	▼47
Nasdaq-100	▼44	▲55	▼40	▼42	▲55	▼40	▲55	▼43	▲60	●52	▼42	▼38	●50	▼45
Dow Jones	▲54	▲54	▲62	▲53	▲56	▲56	●52	●50	▲55	▲55	▲53	●50	▲56	●48
Chip index	▼43	▲54	▼37	▼40	▲60	▼35	▲60	▼42	▲55	▼40	▼40	●48	▼42	▼47
10-year US Treasury yield	▲59	●51	▲62	▲60	▲56	▲65	▲65	▼45	▲55	▲62	▲60	▼40	▲58	▲58
Dollar index	▲55	●51	▲64	▲54	▲55	▼40	▼45	▲56	▲60	▲58	▲54	●52	▲56	▲55
Fear index (VIX)	▼38	▼38	▼35	▼38	▼42	▲65	▼25	●52	▼30	▼44	▼42	▼25	▼40	▲56
Gold	▲56	●51	▲61	▼44	▲55	▲55	▲60	▲57	▲55	●50	▲54	▼42	●52	▼46
WTI crude oil	▲56	▲53	▲54	▼47	▼44	▲57	▲55	▼44	▲60	▼46	▲54	●50	▼44	▲54
Brent crude oil	▲57	▲53	▲54	▼46	▼43	▲56	▲53	▼45	▲60	▼45	▲54	●50	▼44	▲55
Copper	▲57	▲54	▲56	▲58	▲53	▲55	▲60	▼45	▲60	▲58	▲55	●52	●52	▼46
Bitcoin	▼40	●51	▼39	▼47	●48	▼40	▲65	▼44	▲55	●52	▼46	▲58	●49	▼42

Month

	ChatGPT	Claude	Gemini	Grok	Meta AI	Nemotron	Amazon Nova	DeepSeek	Qwen	Kimi	Z.ai	ERNIE	StepFun	MiMo
S&P 500	▲58	▲58	▲60	▼38	▲58	▼45	▼45	●48	▲60	●52	●49	▼40	▼47	▼42
Nasdaq-100	▲55	▲59	●48	▼32	▲58	▼35	▼40	▼45	▲55	▲56	▼40	▼35	▼45	▼38
Dow Jones	▲59	▲56	▲66	▲56	▲57	▲58	▲55	▲57	▲65	▲58	▲56	▲53	▲53	▼46
Chip index	●48	▲58	▼45	▼30	▲62	▼30	▼45	▼46	●50	▼40	▼42	▼42	▼47	▼40
10-year US Treasury yield	▲60	▼44	▲60	▲55	▼44	▲58	▼45	▲56	▲65	▲66	▲62	▼38	▲62	▲62
Dollar index	▲55	▼46	▲63	▲56	●48	▼38	▼40	▲54	▲55	●51	▲58	▲54	▲58	▲53
Fear index (VIX)	▼31	▼28	▼28	●52	▼38	●52	▼20	▼40	▼25	▼46	▼45	▼22	▼46	▲63
Gold	▲60	▲56	▲64	▼40	▲60	▲62	▲55	▲58	●50	▲56	▲55	▲55	●48	▲53
WTI crude oil	▼43	▼46	▼42	▲55	▼40	●51	▼45	●52	▲55	▲55	●48	▼44	▼47	●48
Brent crude oil	▼44	▼46	▼41	▲54	▼39	●50	●48	●52	▲55	▲55	▼47	▼42	●48	●50
Copper	▲61	▲58	▲59	▲62	▲57	▲60	●50	●50	▲65	▲60	▲57	▼40	●50	▼42
Bitcoin	▼47	●51	▲53	▲54	▲54	▲55	▲55	●52	▲65	▲57	▼42	▼42	▼45	▼38

6 months

	ChatGPT	Claude	Gemini	Grok	Meta AI	Nemotron	Amazon Nova	DeepSeek	Qwen	Kimi	Z.ai	ERNIE	StepFun	MiMo
S&P 500	▲64	▲69	▲68	▲55	▲65	▼40	▲65	▲60	▲65	▲60	▲58	▲58	▼42	▲55
Nasdaq-100	▲62	▲70	▲63	●48	▲68	▼30	▲70	▲57	▲70	▲64	▲53	▲55	▼40	●52
Dow Jones	▲64	▲66	▲68	▲62	▲63	●52	▲60	▲60	▲70	▲62	▲58	▲56	●48	▲55
Chip index	▲62	▲66	▲62	●50	▲63	▲55	▲75	▲60	▲65	▲62	▲55	▲58	▲55	●52
10-year US Treasury yield	▼43	▼36	▼38	▼45	▼38	▼42	▼30	▲58	▲70	▲60	▲58	▲65	●48	▼45
Dollar index	▼43	▼30	▼45	▼42	▼42	●52	▼35	▼45	▼45	▼44	▲57	▼42	●48	▼44
Fear index (VIX)	▼29	▼20	▼25	▼45	▼30	▼40	▼15	▼42	▼20	▼35	▼40	▼45	▼40	▼37
Gold	▲64	▲66	▲70	▲54	▲68	▲60	▲70	▲63	▲65	▲62	▲57	▲68	▲56	▲62
WTI crude oil	▼37	▼30	▼35	●52	▼38	▲63	▲60	▲56	▲65	▼45	●52	▲58	▲56	●52
Brent crude oil	▼40	▼28	▼35	●51	▼36	▲62	▲65	▲56	▲65	▼47	●52	▲56	▲56	●51
Copper	▲65	▲66	▲67	▲60	▲62	▲58	▲70	▲58	▲75	▲63	▲61	▲58	▲58	▲53
Bitcoin	▲61	▲64	▲64	▲58	▲60	▲65	▲75	▲58	▲75	▲60	▲53	▲65	▲55	▲55

† — this answer is scored from the price at the moment it was given (the time and price beside the model), to the same close as every answer, in the day’s shared score.

Actions — what the AI would do with its own virtual money: Buy more — would add now; Buy on dips — would add only if the price falls to its level; Hold — would change nothing; Trim — would sell part, keep the rest; Sell — would get out. For the 10-year yield and VIX the AIs bet on the indicator itself.

Ratings of the day

Why did the VIX drop today? 10 fear gauges, Nov 9 2026

Cboe SKEW Index1
Cboe VXN (Nasdaq-100 volatility)2
Cboe VIX1Y and VIX6M (long-dated S&P 500 volatility)3
Cboe VVIX (volatility of the VIX)4
VIX futures curve (November vs December)5

Full ranking

Where the American and the Chinese AIs parted most

Fear index (VIX)1
10-year US Treasury yield2
Nasdaq-1003
Dollar index3
Chip index5

Full ranking

All “Markets” ratings by date

All “Markets” ratings on Top →

S&P 500 · 7,619.38 · Hold · 71% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: 500 of the largest US companies; the main every investment is compared with. Many pension savings and most index funds move with it.

How it counts: the index ^GSPC from the 9:30 AM ET open to the 4:00 PM close. A : a move under 0.073% (a tenth of the day's typical move). The AIs saw the data up to Sat 9:12 AM ET.

Polymarket asks “S&P 500 (SPX) Up or Down on November 9?”: the AIs say Up 54% · people betting real money 54% ($59 traded) · effective voices ≈ 2.3 (mean of the issues). The market ↗ · resolves by wsj.com · price at Sat 9:12 AM ET

S&P 500 (ES) (Thu, 7:54 AM ET): ▼ −0.43%

Blue line: closing prices of our sessions and the latest price. Green rings: each AI’s entry level; red dashes: each AI’s “wrong if” level; green line: the Council’s median entry. Levels more than 12% from the price are not drawn but stay in the AI’s line.

Your call before the AIs: where will S&P 500 go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (7,434.33) to the value at Mon, Nov 9, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Buy on dips · 58% chance of a rise · Target: 7,415.00 · Wrong if: 7,315.00 · Entry: 7,345.00 USA · Gemini 4 Argon · sources: 3

Claude Hold · 54% chance of a rise · Target: 7,395.00 · Wrong if: 7,300.00 USA · Claude Opus 5.5 · sources: 3 · 1 confirm the figure

Meta AI Hold · 47% chance of a rise · Target: 7,358.00 · Wrong if: 7,500.00 USA · Muse Spark 1.3 · sources: 2 · 1 confirm the figure

Grok Hold · 54% chance of a rise · Target: 7,420.00 · Wrong if: 7,280.00 USA · Grok 4.7 · sources: 2

ChatGPT Hold · 55% chance of a rise · Target: 7,402.68 · Wrong if: 7,250.00 USA · GPT-6.1 Sol · sources: 1

Nemotron Buy on dips · 55% chance of a rise · Target: 7,410.00 · Wrong if: 7,320.00 · Entry: 7,340.00 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Hold · 51% chance of a rise · Target: 7,400.00 · Wrong if: 7,300.00 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Buy on dips · 58% chance of a rise · Target: 7,520.00 · Wrong if: 7,180.00 · Entry: 7,280.00 China · Kimi K3 · sources: 3 · 2 confirm the figure

Qwen Buy on dips · 55% chance of a rise · Target: 7,450.00 · Wrong if: 7,250.00 · Entry: 7,300.00 China · Qwen3.8-Max · sources: 1 · 1 confirm the figure

MiMo (Xiaomi) Hold · 55% chance of a rise · Target: 7,420.00 · Wrong if: 7,280.00 China · MiMo-V2.6-Pro

Z.ai Hold · 56% chance of a rise · Target: 7,420.00 · Wrong if: 7,280.00 China · GLM-5.3-Flash · sources: 3

DeepSeek Hold · 52% chance of a rise · Target: 7,420.00 · Wrong if: 7,280.00 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Hold · 54% chance of a rise · Target: 7,410.00 · Wrong if: 7,320.00 China · 文心 5.1 · links opened: 2 of 3 · sources: 3

StepFun Hold · 52% chance of a rise · Target: 7,450.00 · Wrong if: 7,280.00 China · Step 5 Preview · sources: 2

Nasdaq-100 · 28,374.25 · Hold · 57% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: 100 of the largest non-financial companies on Nasdaq — the pulse of big tech and AI. Tech and AI shares in your apps and funds ride on it.

How it counts: the index ^NDX from the 9:30 AM ET open to the 4:00 PM close. A draw: a move under 0.12% (a tenth of the day's typical move). The AIs saw the data up to Sat 9:12 AM ET.

Kalshi, at 4 pm ET above the last close (26,989.99): the AIs say Up 63% · people betting real money 52% (683 contracts traded) · effective voices ≈ 2.3 (mean of the issues). Kalshi ↗ · resolves by google.com · price at Sat 9:12 AM ET

Nasdaq-100 (NQ) (Thu, 7:54 AM ET): ▼ −0.98%

Your call before the AIs: where will Nasdaq-100 go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (27,234.86) to the value at Mon, Nov 9, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Trim · 44% chance of a rise · Target: 26,815.00 · Wrong if: 27,350.00 USA · Gemini 4 Argon · sources: 2

Claude Hold · 54% chance of a rise · Target: 27,100.00 · Wrong if: 26,500.00 USA · Claude Opus 5.5 · sources: 4 · 2 confirm the figure

Meta AI Hold · 46% chance of a rise · Target: 26,882.00 · Wrong if: 27,500.00 USA · Muse Spark 1.3 · sources: 2 · 1 confirm the figure

Grok Trim · 52% chance of a rise · Target: 27,500.00 · Wrong if: 26,200.00 USA · Grok 4.7 · sources: 2

ChatGPT Hold · 56% chance of a rise · Target: 27,165.42 · Wrong if: 26,250.00 USA · GPT-6.1 Sol · sources: 2 · 1 confirm the figure

Nemotron Trim · 42% chance of a rise · Target: 26,900.00 · Wrong if: 27,200.00 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Hold · 52% chance of a rise · Target: 27,100.00 · Wrong if: 26,500.00 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Buy on dips · 57% chance of a rise · Target: 27,500.00 · Wrong if: 25,900.00 · Entry: 26,300.00 China · Kimi K3 · sources: 3 · 1 confirm the figure

Qwen Hold · 55% chance of a rise · Target: 27,200.00 · Wrong if: 26,500.00 China · Qwen3.8-Max · sources: 1

MiMo (Xiaomi) Buy on dips · 58% chance of a rise · Target: 27,400.00 · Wrong if: 26,350.00 · Entry: 26,600.00 China · MiMo-V2.6-Pro

Z.ai Hold · 55% chance of a rise · Target: 27,200.00 · Wrong if: 26,450.00 China · GLM-5.3-Flash · sources: 3 · 1 confirm the figure

DeepSeek Hold · 52% chance of a rise · Target: 27,200.00 · Wrong if: 26,400.00 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Buy on dips · 55% chance of a rise · Target: 27,200.00 · Wrong if: 26,700.00 · Entry: 26,750.00 China · 文心 5.1 · links opened: 1 of 3 · sources: 3

StepFun Hold · 53% chance of a rise · Target: 27,250.00 · Wrong if: 26,700.00 China · Step 5 Preview · sources: 2

Dow Jones · 52,486.49 · Hold · 86% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: 30 giants of different industries; the oldest and best-known index. The headline number the news quotes about the US market.

How it counts: the index ^DJI from the 9:30 AM ET open to the 4:00 PM close. A draw: a move under 0.062% (a tenth of the day's typical move). The AIs saw the data up to Sat 9:12 AM ET.

Dow (YM) (Thu, 7:54 AM ET): ▲ +0.080%

Your call before the AIs: where will Dow Jones go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (51,815.41) to the value at Mon, Nov 9, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Buy on dips · 59% chance of a rise · Target: 51,490.00 · Wrong if: 50,950.00 · Entry: 51,120.00 USA · Gemini 4 Argon · sources: 2

Claude Hold · 53% chance of a rise · Target: 51,360.00 · Wrong if: 50,700.00 USA · Claude Opus 5.5 · sources: 2 · 1 confirm the figure

Meta AI Hold · 52% chance of a rise · Target: 51,388.00 · Wrong if: 50,500.00 USA · Muse Spark 1.3 · sources: 1

Grok Hold · 51% chance of a rise · Target: 51,600.00 · Wrong if: 50,500.00 USA · Grok 4.7 · sources: 1

ChatGPT Hold · 53% chance of a rise · Target: 51,388.30 · Wrong if: 50,400.00 USA · GPT-6.1 Sol · sources: 2

Nemotron Hold · 52% chance of a rise · Target: 51,350.00 · Wrong if: 51,000.00 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Hold · 50% chance of a rise · Target: 51,500.00 · Wrong if: 50,500.00 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Hold · 50% chance of a rise · Target: 51,800.00 · Wrong if: 50,600.00 China · Kimi K3 · sources: 2 · 1 confirm the figure

Qwen Buy on dips · 60% chance of a rise · Target: 51,800.00 · Wrong if: 50,500.00 · Entry: 51,000.00 China · Qwen3.8-Max · sources: 1

MiMo (Xiaomi) Hold · 47% chance of a rise · Target: 51,500.00 · Wrong if: 50,800.00 China · MiMo-V2.6-Pro

Z.ai Hold · 50% chance of a rise · Target: 51,400.00 · Wrong if: 50,700.00 China · GLM-5.3-Flash · sources: 2 · 1 confirm the figure

DeepSeek Hold · 47% chance of a rise · Target: 51,200.00 · Wrong if: 50,600.00 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Hold · 51% chance of a rise · Target: 51,400.00 · Wrong if: 51,000.00 China · 文心 5.1 · sources: 2

StepFun Hold · 50% chance of a rise · Target: 51,600.00 · Wrong if: 50,800.00 China · Step 5 Preview · sources: 2

Chip index · 10,564.11 · Hold · 57% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: Chipmakers: Nvidia, TSMC, ASML, Micron and others. Chips are in every phone, car and AI server — it leads tech.

How it counts: the index ^SOX from the 9:30 AM ET open to the 4:00 PM close. A draw: a move under 0.25% (a tenth of the day's typical move). The AIs saw the data up to Sat 9:12 AM ET.

Gemini Trim · 41% chance of a rise · Target: 9,250.00 · Wrong if: 9,680.00 USA · Gemini 4 Argon · sources: 3 · 1 confirm the figure

Claude Hold · 54% chance of a rise · Target: 9,450.00 · Wrong if: 9,011.00 USA · Claude Opus 5.5 · sources: 3 · 1 confirm the figure

Meta AI Buy on dips · 55% chance of a rise · Target: 9,531.00 · Wrong if: 8,900.00 · Entry: 9,000.00 USA · Muse Spark 1.3 · sources: 2 · 1 confirm the figure

Grok Trim · 55% chance of a rise · Target: 9,600.00 · Wrong if: 9,000.00 USA · Grok 4.7 · sources: 1

ChatGPT Hold · 58% chance of a rise · Target: 9,484.15 · Wrong if: 8,900.00 USA · GPT-6.1 Sol · sources: 2

Nemotron Trim · 38% chance of a rise · Target: 9,280.00 · Wrong if: 9,550.00 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Buy on dips · 55% chance of a rise · Target: 9,500.00 · Wrong if: 9,000.00 · Entry: 9,300.00 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Hold · 47% chance of a rise · Target: 9,750.00 · Wrong if: 9,050.00 China · Kimi K3 · sources: 3 · 2 confirm the figure

Qwen Hold · 45% chance of a rise · Target: 9,500.00 · Wrong if: 9,000.00 China · Qwen3.8-Max · sources: 1

MiMo (Xiaomi) Hold · 53% chance of a rise · Target: 9,600.00 · Wrong if: 9,050.00 China · MiMo-V2.6-Pro

Z.ai Hold · 52% chance of a rise · Target: 9,550.00 · Wrong if: 9,100.00 China · GLM-5.3-Flash · sources: 2 · 2 confirm the figure

DeepSeek Hold · 51% chance of a rise · Target: 9,550.00 · Wrong if: 9,050.00 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Buy more · 62% chance of a rise · Target: 9,600.00 · Wrong if: 9,200.00 China · 文心 5.1 · links opened: 2 of 3 · sources: 2

StepFun Hold · 48% chance of a rise · Target: 9,650.00 · Wrong if: 9,750.00 China · Step 5 Preview · sources: 2

10-year US Treasury yield · 4.998 · Toward a rise, on a dip · 57% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: The most important interest rate in the world: mortgages, loans and stock prices depend on it. Yield up — bonds get cheaper. Your mortgage, loan and savings rates follow it.

Here the AIs bet on the yield itself: “toward a rise” means the yield goes up and bonds get cheaper.

How it counts: the 10-year yield ^TNX, from 9:30 AM to 3:00 PM ET (its last minute), in basis points. A draw: a move under 0.4 bp (a tenth of the day's typical move). The AIs saw the data up to Sat 9:12 AM ET.

Your call before the AIs: where will 10-year US Treasury yield go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (4.835) to the value at Mon, Nov 9, 2026 3:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Toward a rise, on a dip · 58% chance of a rise · Target: 4.890 · Wrong if: 4.790 · Entry: 4.840 USA · Gemini 4 Argon · sources: 3 · 1 confirm the figure

Claude Keep the bet · 50% chance of a rise · Target: 4.860 · Wrong if: 4.950 USA · Claude Opus 5.5 · sources: 4 · 2 confirm the figure

Meta AI Toward a rise, on a dip · 58% chance of a rise · Target: 4.900 · Wrong if: 4.750 · Entry: 4.820 USA · Muse Spark 1.3 · sources: 2 · 2 confirm the figure

Grok Toward a rise, on a dip · 58% chance of a rise · Target: 4.950 · Wrong if: 4.750 · Entry: 4.820 USA · Grok 4.7 · sources: 1

ChatGPT Keep the bet · 55% chance of a rise · Target: 4.890 · Wrong if: 4.780 USA · GPT-6.1 Sol · sources: 2

Nemotron Toward a rise, on a dip · 60% chance of a rise · Target: 4.920 · Wrong if: 4.800 · Entry: 4.830 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Strongly toward a rise · 60% chance of a rise · Target: 4.900 · Wrong if: 4.700 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Toward a rise, on a dip · 60% chance of a rise · Target: 4.940 · Wrong if: 4.760 · Entry: 4.830 China · Kimi K3 · sources: 3 · 2 confirm the figure

Qwen Toward a rise, on a dip · 60% chance of a rise · Target: 4.950 · Wrong if: 4.700 · Entry: 4.800 China · Qwen3.8-Max · sources: 1

MiMo (Xiaomi) Toward a rise, on a dip · 57% chance of a rise · Target: 4.930 · Wrong if: 4.720 · Entry: 4.820 China · MiMo-V2.6-Pro

Z.ai Strongly toward a rise · 57% chance of a rise · Target: 4.900 · Wrong if: 4.780 China · GLM-5.3-Flash · sources: 3 · 1 confirm the figure

DeepSeek Keep the bet · 50% chance of a rise · Target: 4.900 · Wrong if: 4.800 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Toward a fall · 42% chance of a rise · Target: 4.800 · Wrong if: 4.950 China · 文心 5.1 · sources: 2

StepFun Toward a rise, on a dip · 55% chance of a rise · Target: 4.920 · Wrong if: 4.780 · Entry: 4.820 China · Step 5 Preview · sources: 2

Dollar index · 102.92 · Hold · 79% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: The dollar against 6 currencies (euro is 57.6% of the weight). A strong dollar presses on commodities and emerging countries. It sets what your trips, imports and dollar savings cost.

How it counts: the index DX-Y.NYB from the 9:30 AM ET open to the 4:00 PM close. A draw: a move under 0.011% (a tenth of the day's typical move). The AIs saw the data up to Sat 9:12 AM ET.

Your call before the AIs: where will dollar index go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (101.85) to the value at Mon, Nov 9, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Buy on dips · 59% chance of a rise · Target: 102.54 · Wrong if: 101.80 · Entry: 102.10 USA · Gemini 4 Argon · sources: 2

Claude Hold · 50% chance of a rise · Target: 102.30 · Wrong if: 101.60 USA · Claude Opus 5.5 · sources: 3

Meta AI Hold · 52% chance of a rise · Target: 102.43 · Wrong if: 101.00 USA · Muse Spark 1.3 · sources: 1

Grok Hold · 52% chance of a rise · Target: 102.60 · Wrong if: 101.80 USA · Grok 4.7 · sources: 1

ChatGPT Hold · 53% chance of a rise · Target: 102.38 · Wrong if: 101.60 USA · GPT-6.1 Sol · sources: 2

Nemotron Trim · 45% chance of a rise · Target: 102.10 · Wrong if: 102.60 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Hold · 52% chance of a rise · Target: 102.50 · Wrong if: 101.50 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Buy on dips · 56% chance of a rise · Target: 102.75 · Wrong if: 101.60 · Entry: 101.90 China · Kimi K3 · sources: 2

Qwen Hold · 50% chance of a rise · Target: 102.50 · Wrong if: 101.50 China · Qwen3.8-Max · sources: 1

MiMo (Xiaomi) Hold · 54% chance of a rise · Target: 102.70 · Wrong if: 101.70 China · MiMo-V2.6-Pro

Z.ai Hold · 52% chance of a rise · Target: 102.50 · Wrong if: 101.60 China · GLM-5.3-Flash · sources: 2

DeepSeek Hold · 52% chance of a rise · Target: 102.50 · Wrong if: 101.80 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Hold · 53% chance of a rise · Target: 102.50 · Wrong if: 101.50 China · 文心 5.1 · links opened: 2 of 3 · sources: 2

StepFun Hold · 52% chance of a rise · Target: 102.75 · Wrong if: 101.90 China · Step 5 Preview · sources: 2

Fear index (VIX) · 18.35 · Toward a fall · 50% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: Expected market swings for 30 days: below 15 calm, 15–20 normal, 20–30 nervous, above 30 panic. When it jumps, markets are scared — and prices swing hard.

Here the AIs bet on the VIX itself: “toward a rise” means the market gets more nervous.

How it counts: the index ^VIX from the 9:30 AM ET open to the 4:00 PM close. A draw: a move under 0.84% (a tenth of the day's typical move). The AIs saw the data up to Sat 9:12 AM ET.

Your call before the AIs: where will fear index (VIX) go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (23.26) to the value at Mon, Nov 9, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Toward a fall · 40% chance of a rise · Target: 24.48 · Wrong if: 28.20 USA · Gemini 4 Argon · sources: 2 · 1 confirm the figure

Claude Keep the bet · 43% chance of a rise · Target: 25.30 · Wrong if: 29.00 USA · Claude Opus 5.5 · sources: 2 · 1 confirm the figure

Meta AI Toward a rise, on a dip · 55% chance of a rise · Target: 26.67 · Wrong if: 21.00 · Entry: 23.50 USA · Muse Spark 1.3 · sources: 1 · 1 confirm the figure

Grok Toward a fall · 42% chance of a rise · Target: 24.00 · Wrong if: 28.00 USA · Grok 4.7 · sources: 1

ChatGPT Toward a fall · 42% chance of a rise · Target: 24.61 · Wrong if: 29.50 USA · GPT-6.1 Sol · sources: 2

Nemotron Toward a rise, on a dip · 58% chance of a rise · Target: 26.50 · Wrong if: 24.00 · Entry: 25.00 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Toward a fall · 30% chance of a rise · Target: 22.00 · Wrong if: 28.00 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Keep the bet · 48% chance of a rise · Target: 27.00 · Wrong if: 22.50 China · Kimi K3 · sources: 1 · 1 confirm the figure

Qwen Toward a fall · 40% chance of a rise · Target: 24.00 · Wrong if: 28.00 China · Qwen3.8-Max · sources: 1

MiMo (Xiaomi) Keep the bet · 46% chance of a rise · Target: 25.00 · Wrong if: 27.50 China · MiMo-V2.6-Pro

Z.ai Strongly toward a fall · 38% chance of a rise · Target: 23.50 · Wrong if: 29.50 China · GLM-5.3-Flash · sources: 2

DeepSeek Toward a fall · 44% chance of a rise · Target: 24.90 · Wrong if: 28.00 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Strongly toward a fall · 28% chance of a rise · Target: 23.00 · Wrong if: 28.00 China · 文心 5.1 · links opened: 2 of 3 · sources: 2

StepFun Toward a fall · 40% chance of a rise · Target: 21.00 · Wrong if: 28.00 China · Step 5 Preview · sources: 2

Gold · 3,941.10 · Hold · 79% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: A safe-haven asset. Where people hide money when they are worried.

How it counts

How it counts: GC=F — the front-month future (Yahoo rolls it to the next month a few days before expiry), from the 9:30 AM ET price to the 4:00 PM close. A draw: a move under 0.052% (a tenth of the day's typical move). Contract GCZ26 by the exchange calendar, front until Mon, Nov 30, 2026, then GCG27. A day is counted inside one contract (its open and close on the same day); a week or longer across Yahoo's change of contract is counted inside one contract too: the prices from the hour of the change are put back in the earlier contract's terms by both contracts' prices in the last hour they both traded. At the reference minute Yahoo's GC=F stood for GCZ26 (checked hour by hour against the contracts' own prices). The AIs saw the data up to Sat 9:12 AM ET.

Your call before the AIs: where will gold go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (4,136.10) to the value at Mon, Nov 9, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Buy on dips · 57% chance of a rise · Target: 4,093.00 · Wrong if: 4,020.00 · Entry: 4,055.00 USA · Gemini 4 Argon · sources: 1 · 1 confirm the figure

Claude Hold · 51% chance of a rise · Target: 4,080.00 · Wrong if: 3,975.00 USA · Claude Opus 5.5 · sources: 3 · 3 confirm the figure

Meta AI Hold · 48% chance of a rise · Target: 4,062.68 · Wrong if: 4,150.00 USA · Muse Spark 1.3 · sources: 2 · 1 confirm the figure

Grok Hold · 48% chance of a rise · Target: 4,100.00 · Wrong if: 4,000.00 USA · Grok 4.7 · sources: 1 · 1 confirm the figure

ChatGPT Buy on dips · 54% chance of a rise · Target: 4,087.12 · Wrong if: 3,980.00 · Entry: 4,020.00 USA · GPT-6.1 Sol · sources: 2 · 1 confirm the figure

Nemotron Hold · 51% chance of a rise · Target: 4,085.00 · Wrong if: 4,040.00 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Buy on dips · 55% chance of a rise · Target: 4,100.00 · Wrong if: 4,000.00 · Entry: 4,050.00 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Hold · 47% chance of a rise · Target: 4,180.00 · Wrong if: 3,960.00 China · Kimi K3 · sources: 2 · 1 confirm the figure

Qwen Hold · 45% chance of a rise · Target: 4,050.00 · Wrong if: 4,150.00 China · Qwen3.8-Max · sources: 1

MiMo (Xiaomi) Hold · 44% chance of a rise · Target: 4,110.00 · Wrong if: 4,020.00 China · MiMo-V2.6-Pro

Z.ai Hold · 50% chance of a rise · Target: 4,100.00 · Wrong if: 4,020.00 China · GLM-5.3-Flash · sources: 2 · 2 confirm the figure

DeepSeek Hold · 51% chance of a rise · Target: 4,100.00 · Wrong if: 3,980.00 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Hold · 48% chance of a rise · Target: 4,100.00 · Wrong if: 4,030.00 China · 文心 5.1 · links opened: 2 of 3 · sources: 2

StepFun Hold · 50% chance of a rise · Target: 4,130.00 · Wrong if: 3,990.00 China · Step 5 Preview · sources: 2

WTI crude oil · 85.78 · Hold · 57% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: The American oil grade — the price shown in the USA. What you pay for petrol and heating follows it.

How it counts

How it counts: CL=F — the front-month future (Yahoo rolls it to the next month a few days before expiry), from the 9:30 AM ET price to the 4:00 PM close. A draw: a move under 0.13% (a tenth of the day's typical move). Contract CLZ26 by the exchange calendar, front until Fri, Nov 20, 2026, then CLF27. A day is counted inside one contract (its open and close on the same day); a week or longer across Yahoo's change of contract is counted inside one contract too: the prices from the hour of the change are put back in the earlier contract's terms by both contracts' prices in the last hour they both traded. At the reference minute Yahoo's CL=F stood for CLZ26 (checked hour by hour against the contracts' own prices). The AIs saw the data up to Sat 9:12 AM ET.

Polymarket asks “WTI Crude Oil (WTI) Up or Down on November 9?”: the AIs say Up 51% · people betting real money 72% ($42 traded) · effective voices ≈ 2.3 (mean of the issues). The market ↗ · resolves by app.pyth.com · price at Sat 9:12 AM ET · another window: this Polymarket question ends Mon 6:00 PM ET, not at our event's end

Kalshi, at 2:30 PM ET: is the WTI daily settlement (December contract, 2:30 PM ET) above its last Kalshi settlement (90.83)? People betting real money: Up 40% (4,075 contracts traded). The AIs answered Polymarket's question above, not this one. Polymarket and Kalshi differ by 33 percentage points. Kalshi ↗ · resolves by theice.com · price at Sat 9:12 AM ET

Your call before the AIs: where will WTI crude oil go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (86.79) to the value at Mon, Nov 9, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Buy on dips · 57% chance of a rise · Target: 91.70 · Wrong if: 88.40 · Entry: 90.10 USA · Gemini 4 Argon · sources: 3 · 1 confirm the figure

Claude Hold · 56% chance of a rise · Target: 91.50 · Wrong if: 87.50 USA · Claude Opus 5.5 · sources: 3 · 1 confirm the figure

Meta AI Buy on dips · 52% chance of a rise · Target: 91.56 · Wrong if: 87.00 · Entry: 88.00 USA · Muse Spark 1.3 · sources: 2 · 2 confirm the figure

Grok Buy on dips · 51% chance of a rise · Target: 93.00 · Wrong if: 87.00 · Entry: 89.50 USA · Grok 4.7 · sources: 2 · 1 confirm the figure

ChatGPT Buy on dips · 54% chance of a rise · Target: 91.74 · Wrong if: 87.00 · Entry: 89.50 USA · GPT-6.1 Sol · sources: 1

Nemotron Buy on dips · 54% chance of a rise · Target: 91.50 · Wrong if: 89.50 · Entry: 90.20 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Hold · 51% chance of a rise · Target: 91.00 · Wrong if: 89.00 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Hold · 47% chance of a rise · Target: 93.00 · Wrong if: 88.00 China · Kimi K3 · sources: 2 · 1 confirm the figure

Qwen Hold · 50% chance of a rise · Target: 92.00 · Wrong if: 88.00 China · Qwen3.8-Max · sources: 1 · 1 confirm the figure

MiMo (Xiaomi) Hold · 52% chance of a rise · Target: 91.50 · Wrong if: 89.50 China · MiMo-V2.6-Pro

Z.ai Hold · 55% chance of a rise · Target: 91.60 · Wrong if: 88.50 China · GLM-5.3-Flash · sources: 3 · 2 confirm the figure

DeepSeek Hold · 45% chance of a rise · Target: 90.00 · Wrong if: 94.00 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Buy on dips · 52% chance of a rise · Target: 91.50 · Wrong if: 89.50 · Entry: 90.00 China · 文心 5.1 · links opened: 1 of 3 · sources: 2

StepFun Hold · 46% chance of a rise · Target: 91.50 · Wrong if: 93.00 China · Step 5 Preview · sources: 2

Brent crude oil · 92.11 · Hold · 57% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: The world benchmark: most of the world's oil is priced by it. Most of the world's fuel prices follow it.

How it counts

How it counts: BZ=F — the front-month future (Yahoo rolls it to the next month a few days before expiry), from the 9:30 AM ET price to the 4:00 PM close. A draw: a move under 0.15% (a tenth of the day's typical move). Contract BZF27 by the exchange calendar, front until Mon, Nov 30, 2026, then BZG27. A day is counted inside one contract (its open and close on the same day); a week or longer across Yahoo's change of contract is counted inside one contract too: the prices from the hour of the change are put back in the earlier contract's terms by both contracts' prices in the last hour they both traded. At the reference minute Yahoo's BZ=F stood for BZF27 (checked hour by hour against the contracts' own prices). The AIs saw the data up to Sat 9:12 AM ET.

Your call before the AIs: where will Brent crude oil go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (93.38) to the value at Mon, Nov 9, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Buy on dips · 56% chance of a rise · Target: 99.20 · Wrong if: 95.80 · Entry: 97.50 USA · Gemini 4 Argon · sources: 2

Claude Hold · 55% chance of a rise · Target: 98.90 · Wrong if: 95.00 USA · Claude Opus 5.5 · sources: 4 · 1 confirm the figure

Meta AI Buy on dips · 52% chance of a rise · Target: 99.01 · Wrong if: 94.00 · Entry: 95.00 USA · Muse Spark 1.3 · sources: 2 · 1 confirm the figure

Grok Buy on dips · 50% chance of a rise · Target: 100.50 · Wrong if: 94.00 · Entry: 96.50 USA · Grok 4.7 · sources: 1 · 1 confirm the figure

ChatGPT Buy on dips · 55% chance of a rise · Target: 99.70 · Wrong if: 94.00 · Entry: 96.50 USA · GPT-6.1 Sol · sources: 1

Nemotron Buy on dips · 53% chance of a rise · Target: 98.80 · Wrong if: 97.00 · Entry: 97.80 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Hold · 50% chance of a rise · Target: 98.50 · Wrong if: 97.50 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Hold · 47% chance of a rise · Target: 100.50 · Wrong if: 95.50 China · Kimi K3 · sources: 2

Qwen Hold · 50% chance of a rise · Target: 99.50 · Wrong if: 96.00 China · Qwen3.8-Max · sources: 1 · 1 confirm the figure

MiMo (Xiaomi) Hold · 51% chance of a rise · Target: 99.00 · Wrong if: 97.00 China · MiMo-V2.6-Pro

Z.ai Hold · 54% chance of a rise · Target: 99.00 · Wrong if: 96.20 China · GLM-5.3-Flash · sources: 3 · 2 confirm the figure

DeepSeek Hold · 45% chance of a rise · Target: 97.50 · Wrong if: 101.50 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Buy on dips · 53% chance of a rise · Target: 99.00 · Wrong if: 97.00 · Entry: 97.50 China · 文心 5.1 · links opened: 2 of 3 · sources: 2

StepFun Hold · 46% chance of a rise · Target: 99.00 · Wrong if: 100.50 China · Step 5 Preview · sources: 2

Copper · 6.202 · Hold · 71% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: Construction, factories, China: copper rises when the economy speeds up. A read of how fast factories and building are going.

How it counts

How it counts: HG=F — the front-month future (Yahoo rolls it to the next month a few days before expiry), from the 9:30 AM ET price to the 4:00 PM close. A draw: a move under 0.063% (a tenth of the day's typical move). Contract HGZ26 by the exchange calendar, front until Mon, Nov 30, 2026, then HGH27. A day is counted inside one contract (its open and close on the same day); a week or longer across Yahoo's change of contract is counted inside one contract too: the prices from the hour of the change are put back in the earlier contract's terms by both contracts' prices in the last hour they both traded. At the reference minute Yahoo's HG=F stood for HGZ26 (checked hour by hour against the contracts' own prices). The AIs saw the data up to Sat 9:12 AM ET.

Your call before the AIs: where will copper go on Nov 9, 2026?

Counted from the value at Mon, Nov 9, 2026 9:30 AM ET (6.326) to the value at Mon, Nov 9, 2026 4:00 PM ET: higher at the end — up, lower — down; a move under the draw threshold is a draw.

Gemini Hold · 52% chance of a rise · Target: 6.260 · Wrong if: 6.120 USA · Gemini 4 Argon · sources: 2

Claude Hold · 52% chance of a rise · Target: 6.250 · Wrong if: 6.080 USA · Claude Opus 5.5 · sources: 5 · 5 confirm the figure

Meta AI Hold · 49% chance of a rise · Target: 6.230 · Wrong if: 6.380 USA · Muse Spark 1.3 · sources: 1 · 1 confirm the figure

Grok Buy on dips · 56% chance of a rise · Target: 6.350 · Wrong if: 6.050 · Entry: 6.150 USA · Grok 4.7 · sources: 2 · 1 confirm the figure

ChatGPT Buy on dips · 55% chance of a rise · Target: 6.260 · Wrong if: 6.080 · Entry: 6.180 USA · GPT-6.1 Sol · sources: 2

Nemotron Hold · 51% chance of a rise · Target: 6.280 · Wrong if: 6.180 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova Buy on dips · 55% chance of a rise · Target: 6.300 · Wrong if: 6.100 · Entry: 6.200 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi Buy on dips · 56% chance of a rise · Target: 6.420 · Wrong if: 6.040 · Entry: 6.150 China · Kimi K3 · sources: 3 · 1 confirm the figure

Qwen Hold · 45% chance of a rise · Target: 6.200 · Wrong if: 6.400 China · Qwen3.8-Max · sources: 2

MiMo (Xiaomi) Hold · 48% chance of a rise · Target: 6.300 · Wrong if: 6.120 China · MiMo-V2.6-Pro

Z.ai Hold · 52% chance of a rise · Target: 6.280 · Wrong if: 6.120 China · GLM-5.3-Flash · sources: 1

DeepSeek Hold · 48% chance of a rise · Target: 6.220 · Wrong if: 6.400 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin) Hold · 47% chance of a rise · Target: 6.300 · Wrong if: 6.100 China · 文心 5.1 · sources: 2

StepFun Hold · 48% chance of a rise · Target: 6.320 · Wrong if: 6.120 China · Step 5 Preview · sources: 2

Bitcoin · 72,794 · Hold · 57% of the AIs chose it

Ask the Council about this market — coming to the Poly app → a public question: one sealed answer for everyone who asks about that ticker that day

What it is: Trades around the clock, so it reacts first to weekend news. The 24/7 mood of risk — it moves first at the weekend.

How it counts: BTC-USD — counted from the price at the seal (Sat 10:23 AM ET) to Sun 10:23 AM ET. A draw: a move under 0.054% (a tenth of the day's typical move). The AIs saw the data up to Sat 9:12 AM ET.

Your call before the AIs: where will Bitcoin go on Nov 8, 2026?

Counted from the price at the seal, $75,773 (Sat 10:23 AM ET), to Sun 10:23 AM ET.

Gemini† answer 9:18 AM ET · from 75,773 Trim · 43% chance of a rise · Target: 75,150 · Wrong if: 77,400 USA · Gemini 4 Argon · sources: 3 · 3 confirm the figure

Claude† answer 9:55 AM ET · from 75,773 Hold · 52% chance of a rise · Target: 76,000 · Wrong if: 74,000 USA · Claude Opus 5.5 · sources: 4 · 4 confirm the figure

Meta AI† answer 10:22 AM ET · from 75,773 Hold · 47% chance of a rise · Target: 75,456 · Wrong if: 77,500 USA · Muse Spark 1.3 · sources: 2 · 1 confirm the figure

Grok† answer 9:16 AM ET · from 75,773 Hold · 51% chance of a rise · Target: 77,000 · Wrong if: 73,500 USA · Grok 4.7 · sources: 1

ChatGPT† answer 9:24 AM ET · from 75,773 Trim · 45% chance of a rise · Target: 75,152 · Wrong if: 78,000 USA · GPT-6.1 Sol · sources: 2

Nemotron† answer 9:15 AM ET · from 75,773 Trim · 45% chance of a rise · Target: 75,200 · Wrong if: 77,000 USA · Nemotron 3 Ultra · sources: 3

Amazon Nova† answer 9:14 AM ET · from 75,773 Buy on dips · 60% chance of a rise · Target: 76,000 · Wrong if: 73,000 · Entry: 74,000 USA · Amazon Nova · links opened: 0 of 1 · sources: 1

Kimi† answer 9:23 AM ET · from 75,773 Buy on dips · 56% chance of a rise · Target: 78,500 · Wrong if: 73,000 · Entry: 74,500 China · Kimi K3 · sources: 3 · 1 confirm the figure

Qwen† answer 9:19 AM ET · from 75,773 Buy on dips · 55% chance of a rise · Target: 77,000 · Wrong if: 72,000 · Entry: 74,000 China · Qwen3.8-Max · sources: 1

MiMo (Xiaomi)† answer 10:33 AM ET · from 75,697 Hold · 45% chance of a rise · Target: 76,500 · Wrong if: 73,800 China · MiMo-V2.6-Pro

Z.ai† answer 9:32 AM ET · from 75,773 Hold · 47% chance of a rise · Target: 74,500 · Wrong if: 78,500 China · GLM-5.3-Flash · sources: 2 · 1 confirm the figure

DeepSeek† answer 9:16 AM ET · from 75,773 Hold · 48% chance of a rise · Target: 76,500 · Wrong if: 73,500 China · DeepSeek V4.1 · sources: 2

ERNIE (Wenxin)† answer 9:17 AM ET · from 75,773 Hold · 55% chance of a rise · Target: 76,500 · Wrong if: 74,800 China · 文心 5.1 · sources: 2

StepFun† answer 9:21 AM ET · from 75,773 Hold · 50% chance of a rise · Target: 77,000 · Wrong if: 73,500 China · Step 5 Preview · sources: 2

The league’s full table, with the Pro league →

† — this answer is scored from the price at the moment it was given (the time and price beside the model), to the same close as every answer, in the day’s shared score.

The league of virtual accounts: the AIs and the crowd side by side

League summary as of the session of Nov 11, 2026.

The league. Session Wed, Nov 11, 2026: traded — DeepSeek, Kimi · answered and kept the weights — ChatGPT, MiMo (Xiaomi), Gemini, Z.ai, Grok, Claude, Qwen, Meta AI, StepFun · positions unchanged — Nemotron, Amazon Nova, ERNIE (Wenxin).

Marks: ▲ up · ▼ down · ● even (the heat map); ● traded at the session · ○ kept its weights (the league).

Every AI keeps three separate virtual $100,000 accounts, so one AI has three figures: the League (buying only), the Pro league ( and leverage up to 2×) and the portfolio by actions (from its calls on the markets).

Both leagues in short: each AI’s money and status

League^‡

Money: each AI manages a virtual $100,000, buying only — who has made more.

Starting line, not a ranking: no AI has 30 closed forecasts yet.

Participant	Equity	Change	Status
S&P 500, bought once and held since Mon, Oct 19 · from the 10-19 open, first price 09:30–09:45 , after costs	$98,885	−$1,115−1.12%	—
1 ChatGPTUSA · Council	$98,740	−$1,260−1.26%	○ kept the weights
2 MiMo (Xiaomi)China · Council	$98,317	−$1,683−1.68%	○ kept the weights · new decision accepted, takes effect Thu, Nov 12
3 StepFunChina · bench	$98,018	−$1,982−1.98%	○ kept the weights
4 GeminiUSA · Council	$97,982	−$2,018−2.02%	○ kept the weights
5 Z.aiChina · Council	$97,812	−$2,188−2.19%	○ kept the weights
6 NemotronUSA · bench	$97,751	−$2,249−2.25%	○ holds its positions
7 Amazon NovaUSA · bench	$96,820	−$3,180−3.18%	○ holds its positions · new decision accepted, takes effect Thu, Nov 12
8 GrokUSA · Council	$96,818	−$3,182−3.18%	○ kept the weights
9 DeepSeekChina · Council	$96,658	−$3,342−3.34%	● traded: 7 trades
10 ClaudeUSA · Council	$96,583	−$3,417−3.42%	○ kept the weights
11 KimiChina · Council	$96,581	−$3,419−3.42%	● traded: 7 trades
12 QwenChina · Council	$95,631	−$4,369−4.37%	○ kept the weights
13 Meta AIUSA · Council	$95,287	−$4,713−4.71%	○ kept the weights
14 ERNIE (Wenxin)China · bench	$94,225	−$5,775−5.78%	○ holds its positions · new decision accepted, takes effect Thu, Nov 12
More rows — the whole table ↓

Pro league^‡

the same AIs and money, plus shorts and up to 2× leverage · leader Gemini +$2,922 (+2.92%) · S&P 500 since Tue, Oct 20 −0.85%

Participant	Equity	Change	Status
1 Gemini	$102,922	+$2,922+2.92%	○ kept the weights
S&P 500, bought once and held since Tue, Oct 20 · from the 10-20 open, first price 09:30–09:45 ET, after costs	$99,154	−$846−0.85%	—
2 StepFun	$98,404	−$1,596−1.60%	○ kept the weights
3 MiMo (Xiaomi)	$97,596	−$2,404−2.40%	○ kept the weights · new decision accepted, takes effect Thu, Nov 12
4 DeepSeek	$97,418	−$2,582−2.58%	● traded: 9 trades
5 Nemotron	$97,238	−$2,762−2.76%	○ holds its positions
6 ChatGPT	$97,225	−$2,775−2.78%	● traded: 6 trades
7 Grok	$96,626	−$3,374−3.37%	○ kept the weights
8 Z.ai	$96,377	−$3,623−3.62%	○ kept the weights
9 Amazon Nova	$95,686	−$4,314−4.31%	● traded: 6 trades
10 Claude	$95,168	−$4,832−4.83%	○ kept the weights
11 Qwen	$94,457	−$5,543−5.54%	● traded: 8 trades
12 ERNIE (Wenxin)	$94,351	−$5,649−5.65%	● traded: 9 trades
13 Meta AI	$92,760	−$7,240−7.24%	○ kept the weights
14 Kimi	$92,105	−$7,895−7.90%	○ kept the weights

Rating

collecting history: 30 of 30 resolved forecasts · trading days: 10/10 Full rating →

The rule of the game: to tell 55% right calls from a coin takes about 209 sessions ≈ 10 months; the ranking comes earlier and sharpens every day. (calculated assuming the forecasts of one day are linked, ρ≈0.2; from 5 sessions — with the link measured in the book)

Coverage and Brier for each AI ▾

In each cell: how many of the questions asked the AI answered with a probability; then its Brier (0 is perfect, 0.25 is a coin) on its own answers, and with every gap counted as a 50% answer. A 50% the AI said itself and a gap counted as 50% are told apart.

Who	Day	Week	Month	6 months
ChatGPT	answered 89% Brier 0.247 on its own answers, 0.247 with the gaps counted as 50%	answered 89% Brier 0.249 on its own answers, 0.249 with the gaps counted as 50% · said “50%” itself: 1	answered 89% no outcome yet — the horizon has not ended	answered 89% no outcome yet — the horizon has not ended
Claude	answered 95% Brier 0.247 on its own answers, 0.247 with the gaps counted as 50% · said “50%” itself: 30	answered 95% Brier 0.259 on its own answers, 0.258 with the gaps counted as 50% · said “50%” itself: 24	answered 95% no outcome yet — the horizon has not ended · said “50%” itself: 6	answered 95% no outcome yet — the horizon has not ended · said “50%” itself: 2
Gemini	answered 89% Brier 0.251 on its own answers, 0.251 with the gaps counted as 50% · said “50%” itself: 1	answered 89% Brier 0.251 on its own answers, 0.251 with the gaps counted as 50%	answered 89% no outcome yet — the horizon has not ended · said “50%” itself: 1	answered 89% no outcome yet — the horizon has not ended
Grok	answered 95% Brier 0.239 on its own answers, 0.240 with the gaps counted as 50% · said “50%” itself: 5	answered 95% Brier 0.258 on its own answers, 0.258 with the gaps counted as 50% · said “50%” itself: 6	answered 95% no outcome yet — the horizon has not ended · said “50%” itself: 19	answered 95% no outcome yet — the horizon has not ended · said “50%” itself: 10
Meta AI	answered 93% Brier 0.239 on its own answers, 0.239 with the gaps counted as 50% · said “50%” itself: 2	answered 93% Brier 0.256 on its own answers, 0.256 with the gaps counted as 50% · said “50%” itself: 4	answered 93% no outcome yet — the horizon has not ended · said “50%” itself: 3	answered 93% no outcome yet — the horizon has not ended · said “50%” itself: 1
Nemotron	answered 84% Brier 0.259 on its own answers, 0.258 with the gaps counted as 50% · said “50%” itself: 6 · gaps counted as 50%: 4	answered 84% Brier 0.251 on its own answers, 0.251 with the gaps counted as 50% · said “50%” itself: 9 · gaps counted as 50%: 4	answered 84% no outcome yet — the horizon has not ended · said “50%” itself: 15	answered 84% no outcome yet — the horizon has not ended · said “50%” itself: 3
Amazon Nova	answered 80% Brier 0.242 on its own answers, 0.243 with the gaps counted as 50% · said “50%” itself: 14	answered 80% Brier 0.326 on its own answers, 0.316 with the gaps counted as 50% · said “50%” itself: 17	answered 80% no outcome yet — the horizon has not ended · said “50%” itself: 33	answered 80% no outcome yet — the horizon has not ended · said “50%” itself: 20
DeepSeek	answered 89% Brier 0.239 on its own answers, 0.240 with the gaps counted as 50% · said “50%” itself: 12	answered 89% Brier 0.253 on its own answers, 0.253 with the gaps counted as 50% · said “50%” itself: 9	answered 89% no outcome yet — the horizon has not ended · said “50%” itself: 10	answered 89% no outcome yet — the horizon has not ended · said “50%” itself: 6
Qwen	answered 81% Brier 0.236 on its own answers, 0.238 with the gaps counted as 50% · said “50%” itself: 22	answered 81% Brier 0.292 on its own answers, 0.291 with the gaps counted as 50% · said “50%” itself: 7	answered 81% no outcome yet — the horizon has not ended · said “50%” itself: 12	answered 81% no outcome yet — the horizon has not ended · said “50%” itself: 12
Kimi	answered 83% Brier 0.251 on its own answers, 0.251 with the gaps counted as 50% · said “50%” itself: 6	answered 83% Brier 0.266 on its own answers, 0.265 with the gaps counted as 50% · said “50%” itself: 8	answered 83% no outcome yet — the horizon has not ended · said “50%” itself: 3	answered 83% no outcome yet — the horizon has not ended · said “50%” itself: 8
Z.ai	answered 94% Brier 0.241 on its own answers, 0.241 with the gaps counted as 50% · said “50%” itself: 6	answered 94% Brier 0.266 on its own answers, 0.265 with the gaps counted as 50% · said “50%” itself: 9	answered 94% no outcome yet — the horizon has not ended · said “50%” itself: 7	answered 94% no outcome yet — the horizon has not ended · said “50%” itself: 10
ERNIE (Wenxin)	answered 76% Brier 0.252 on its own answers, 0.251 with the gaps counted as 50% · said “50%” itself: 18	answered 76% Brier 0.232 on its own answers, 0.233 with the gaps counted as 50% · said “50%” itself: 9	answered 76% no outcome yet — the horizon has not ended · said “50%” itself: 9	answered 76% no outcome yet — the horizon has not ended · said “50%” itself: 9
StepFun	answered 94% Brier 0.242 on its own answers, 0.243 with the gaps counted as 50% · said “50%” itself: 27	answered 94% Brier 0.249 on its own answers, 0.249 with the gaps counted as 50% · said “50%” itself: 15	answered 94% no outcome yet — the horizon has not ended · said “50%” itself: 8	answered 94% no outcome yet — the horizon has not ended · said “50%” itself: 18
MiMo (Xiaomi)	answered 95% Brier 0.252 on its own answers, 0.252 with the gaps counted as 50% · said “50%” itself: 6	answered 95% Brier 0.243 on its own answers, 0.243 with the gaps counted as 50% · said “50%” itself: 11	answered 95% no outcome yet — the horizon has not ended · said “50%” itself: 5	answered 95% no outcome yet — the horizon has not ended · said “50%” itself: 3

AI Portfolio League^‡

Virtual portfolios

Every model gets the same simulated $100,000, the same data and the same rules — and manages it itself: a full review once a week, a change any day it sees a reason. Not who says the direction right, but who keeps and grows the capital for the risk.

One reading for the 10-year yield and VIX: the day’s call is a bet on the indicator itself (buy = the yield or the VIX rises); in the portfolio leagues the 10-year is the IEF fund, held long, and it rises when the yield falls.

A public audit of how AI models turn forecasts, confidence and disagreement into managing capital. How this differs from other AI trading contests: the leading AIs of the US and of China, 12 markets, every forecast and every decision before it starts to count, two leagues — long-only and Pro with and leverage — and all the data open to check and download.

Long positions and cash only, no leverage, 5–25% a position — or “hold”: the positions stay, no trade and no cost. Every day each model may change its portfolio (a full review once a week, on the week's first session); a decision is sealed and trades at the first open after it, at the first price of 09:30–09:45 , 0.1% on what changes; the program turns weights into trades and publishes each one; the cash earns the T-bill. VIX is an indicator only — it cannot be held. Each model gets its own account in its question (never the others' weights). The account runs on; a season (126 sessions) is a slice of it. Units are fractional, like fractional shares at a broker: any amount buys the exact weight. Like a football squad: each day the coach — the — sends the starting line-up onto the field; the rest sit on the bench, and they win or lose with the team: their positions keep working. It is a contest of systems: each AI works with its own search at the mode named in its Configuration line. The 0.1% is one rate for every cost of a changed dollar; a fund's dividend counts from its ex-date and becomes cash about 10 days later.

Decisions any day by the published rule (arena-protocol.json): the weekly review, the radar and the league guardian — a model with no accepted decision this week is asked again hourly 07:00–21:00 ET, at most 5 times a day; every decision trades at the next open; the Council’s position changes only in the scheduled issue. Counted in the leagues (arena-decisions.csv): the league guardian — main league 1, Pro league 2; the radar — main league 4, Pro league 2.

The league's protocol: version 3 and its history

version 1 (Sun, Oct 18, 2026): the first rules: long only, 5–25% a position, the rest cash at the T-bill, 0.1% on what changes, the first price of 09:30–09:45 ET.
version 2 (Mon, Oct 19, 2026): “hold” is a decision with no trade and no cost; each model sees only its own account; VIX became an indicator only; a decision any day.
version 3 (Tue, Oct 20, 2026): a fund may name its market (SPY = S&P 500); the model states its mandate; the positions and cash must sum to 100 within 0.5. The same weights again bring back a position that drifted 2 p.p. or more.

The full protocol, the question's template and its schema (JSON)

Who would you trust with a virtual $100,000?

See it on your own amount: $^‡

The day 2026-11-11: best — MiMo (Xiaomi) +$1,203 (+1.24%) · smallest gain — ERNIE (Wenxin) +$82 (+0.09%)

MiMo (Xiaomi) made the most on Chip index (+$869) · ERNIE (Wenxin) lost the most on Gold (−$133) · They bet apart on Brent crude oil: ERNIE (Wenxin) held 19%, MiMo (Xiaomi) none

Overtakes of the day: MiMo (Xiaomi) passed Gemini; MiMo (Xiaomi) passed StepFun; MiMo (Xiaomi) passed Z.ai; Amazon Nova passed Claude; Amazon Nova passed DeepSeek; Amazon Nova passed Kimi; DeepSeek passed Kimi

Participant	Value	Week	Return	vs S&P 500, p.p.	Max drawdown	Return per risk	Cash, avg	Real diversification	Turnover	Last action
ChatGPT	$98,740	+$207 +0.21%	−1.26%	−0.15	−2.00%	—	33%	4.2	1.11	held on purpose Wed, Nov 11, 2026 decisions: 4 of 4 weeks
MiMo (Xiaomi)	$98,317	+$2,645 +2.76%	−1.68%	−0.57	−4.73%	—	30%	6.0	1.60	held on purpose Wed, Nov 11, 2026 new decision accepted, takes effect Thu, Nov 12, 2026 decisions: 4 of 4 weeks
StepFun	$98,018	+$597 +0.61%	−1.98%	−0.87	−2.90%	—	28%	4.8	1.63	held on purpose Wed, Nov 11, 2026 decisions: 4 of 4 weeks
Gemini	$97,982	+$238 +0.24%	−2.02%	−0.90	−3.04%	—	21%	4.6	0.94	held on purpose Wed, Nov 11, 2026 decisions: 4 of 4 weeks
Z.ai	$97,812	−$59 −0.06%	−2.19%	−1.07	−2.91%	—	52%	4.9	1.20	held on purpose Wed, Nov 11, 2026 decisions: 4 of 4 weeks
Nemotron	$97,751	+$1,719 +1.79%	−2.25%	−1.13	−4.94%	—	26%	7.9	1.80	holds its positions Wed, Nov 11, 2026 decisions: 4 of 4 weeks
Amazon Nova	$96,820	+$1,310 +1.37%	−3.18%	−2.06	−5.12%	—	24%	6.7	1.61	holds its positions Wed, Nov 11, 2026 new decision accepted, takes effect Thu, Nov 12, 2026 decisions: 4 of 4 weeks
Grok	$96,818	+$142 +0.15%	−3.18%	−2.07	−3.74%	—	28%	5.1	0.93	held on purpose Wed, Nov 11, 2026 decisions: 4 of 4 weeks
DeepSeek	$96,658	+$257 +0.27%	−3.34%	−2.23	−4.05%	—	13%	5.6	1.32	traded: 7 trades Wed, Nov 11, 2026 decisions: 4 of 4 weeks
Claude	$96,583	+$231 +0.24%	−3.42%	−2.30	−4.29%	—	21%	5.6	1.21	held on purpose Wed, Nov 11, 2026 decisions: 4 of 4 weeks
Kimi	$96,581	+$84 +0.09%	−3.42%	−2.30	−4.25%	—	21%	3.9	2.47	traded: 7 trades Wed, Nov 11, 2026 decisions: 4 of 4 weeks
Qwen	$95,631	+$87 +0.09%	−4.37%	−3.25	−6.03%	—	19%	6.8	3.96	held on purpose Wed, Nov 11, 2026 decisions: 4 of 4 weeks
Meta AI	$95,287	+$563 +0.59%	−4.71%	−3.60	−6.42%	—	15%	5.3	3.67	held on purpose Wed, Nov 11, 2026 decisions: 4 of 4 weeks
ERNIE (Wenxin)	$94,225	−$1,695 −1.77%	−5.78%	−4.66	−7.57%	—	22%	5.5	1.06	holds its positions Wed, Nov 11, 2026 new decision accepted, takes effect Thu, Nov 12, 2026 decisions: 3 of 4 weeks
Yardsticks to compare with
S&P 500, bought once and held · from the 10-19 open, first price 09:30–09:45 ET, after costs	$98,885	+$3,191 +3.33%	−1.12%	—	−5.19%	—	0%	1.0	1.00	held on purpose Mon, Nov 9, 2026 decisions: 4 of 4 weeks
Cash only	$100,258	+$56 +0.06%	+0.26%	+1.37	0.00%	—	100%	—	0.00	kept the weights Mon, Nov 9, 2026 decisions: 4 of 4 weeks
The readers (the median of the readers' own portfolios, under the same rules)	$100,258	+$56 +0.06%	+0.26%	+1.37	0.00%	—	100%	—	0.00	positions unchanged decisions: 0 of 4 weeks
60/40 stocks/bonds (outside yardstick, above the 25% cap)	$99,594	+$1,556 +1.59%	−0.41%	+0.71	−2.59%	—	0%	1.9	1.03	traded: 2 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks
The Council by confidence (each weight times the AI's own confidence)	$99,382	+$357 +0.36%	−0.62%	+0.50	−1.12%	—	70%	3.0	0.65	traded: 4 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks
The Council with its disagreement (more disagreement — more cash)	$98,907	+$377 +0.38%	−1.09%	+0.02	−1.77%	—	59%	3.8	0.76	traded: 4 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks
Team USA (the US models' mean weights)	$97,700	+$218 +0.22%	−2.30%	−1.18	−2.88%	—	34%	5.9	1.31	traded: 9 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks
The Council (mean of the AIs' weights)	$97,069	−$43 −0.04%	−2.93%	−1.82	−3.38%	—	42%	3.8	1.35	traded: 7 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks
Team China (the Chinese models' mean weights)	$96,536	−$2 +0.00%	−3.46%	−2.35	−3.97%	—	35%	3.9	1.71	traded: 7 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks
The Council by past accuracy (each AI weighted by its right calls)	$96,513	+$160 +0.17%	−3.49%	−2.37	−4.08%	—	36%	4.7	1.33	traded: 8 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks
Where 3+ models agree (those markets alike)	$95,948	+$2,873 +3.09%	−4.05%	−2.94	−8.53%	—	0%	7.0	3.13	kept the weights Wed, Nov 11, 2026 decisions: 4 of 4 weeks
Equal weight	$95,480	+$839 +0.89%	−4.52%	−3.40	−6.99%	—	0%	11.0	1.08	traded: 11 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks
All traded markets, bought once and held	$95,327	+$686 +0.73%	−4.67%	−3.56	−6.93%	—	0%	11.0	1.00	held on purpose Mon, Nov 9, 2026 decisions: 4 of 4 weeks
Where 1–2 models are (the rest alike)	$95,066	+$1,381 +1.47%	−4.93%	−3.82	−7.66%	—	3%	5.0	10.80	traded: 7 trades Wed, Nov 11, 2026 decisions: 4 of 4 weeks
The Council by majority (markets most AIs hold, alike)	$94,600	−$143 −0.15%	−5.40%	−4.29	−6.29%	—	0%	4.0	2.37	traded: 5 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks
Random within the rules	$94,338	+$253 +0.27%	−5.66%	−4.55	−7.22%	—	9%	5.7	4.25	traded: 10 trades Mon, Nov 9, 2026 decisions: 4 of 4 weeks

The market over the same period: S&P 500 −1.12%, Bitcoin −9.51%, “equal” −4.52%. The market fell. The whole field's day return moves with the S&P 500 at a correlation of 0.89 — a lean, not skill.

“Return per risk” is the return above the T-bill per unit of the account's swings — professionals call it the Sharpe ratio. The main league trades 11 of the 12 markets: VIX only in the (through the VIXY fund). What each account trades: S&P 500 — SPY, Nasdaq-100 — QQQ, Dow Jones — DIA, Chip index — SOXQ, US 10-year bonds (IEF — rises when the yield falls), Dollar index — UUP, Gold — GLD, WTI crude oil — USO, Brent crude oil — BNO, Copper — CPER, Bitcoin — BTC-USD.

Overtakes: Wed, Nov 11, 2026 MiMo (Xiaomi) passed Gemini; Wed, Nov 11, 2026 MiMo (Xiaomi) passed StepFun; Wed, Nov 11, 2026 MiMo (Xiaomi) passed Z.ai; Wed, Nov 11, 2026 Amazon Nova passed Claude; Wed, Nov 11, 2026 Amazon Nova passed DeepSeek; Wed, Nov 11, 2026 Amazon Nova passed Kimi; Wed, Nov 11, 2026 DeepSeek passed Kimi; Tue, Nov 10, 2026 StepFun passed Gemini

Right more often ≠ earned more

The same period — the league's sessions since 2026-10-19. On the left — the share of right day calls with a side; on the right — the account's return.

Right more often

#	Participant	Right	Calls	Place by money minus place here
1	Z.ai	61%	168	+4
2	Amazon Nova	61%	107	+5
3	Qwen	60%	121	+9
4	Grok	58%	192	+3
5	DeepSeek	56%	165	+4
6	StepFun	56%	139	−3
7	Meta AI	56%	177	+6
8	Claude	53%	170	+2
9	Kimi	53%	157	+1
10	ChatGPT	53%	173	−9
11	Gemini	52%	183	−7
12	MiMo (Xiaomi)	52%	190	−10
13	Nemotron	52%	133	−7
14	ERNIE (Wenxin)	51%	111	+0

Earned more

#	Participant	Return
1	ChatGPT	−1.26%
2	MiMo (Xiaomi)	−1.68%
3	StepFun	−1.98%
4	Gemini	−2.02%
5	Z.ai	−2.19%
6	Nemotron	−2.25%
7	Amazon Nova	−3.18%
7	Grok	−3.18%
9	DeepSeek	−3.34%
10	Claude	−3.42%
10	Kimi	−3.42%
12	Qwen	−4.37%
13	Meta AI	−4.71%
14	ERNIE (Wenxin)	−5.78%

The Council −2.93% against the mean model −3.06% and the best one, ChatGPT, −1.26% — the same period.

The week and the season in a few words

The week since 2026-11-09: best MiMo (Xiaomi) +2.76%, worst ERNIE (Wenxin) −1.77%; “equal” +0.89%, S&P 500 +3.33%; all the AIs' costs this week $313. MiMo (Xiaomi) made the most on Chip index (+$1,892). Since the season began: ahead ChatGPT −1.26% (deepest fall 2.00%), behind ERNIE (Wenxin) −5.78%; “equal” −4.52%, S&P 500 −1.12%. A story, not a verdict.

Your own virtual $100,000 — against the AIs

The same rules: each position 5–25%, the rest in cash. Your weights are sealed by the server's clock and trade at the next open, like the AIs'.

S&P 500 (SPY) % Nasdaq-100 (QQQ) % Dow Jones (DIA) % Chip index (SOXQ) % US 10-year bonds (IEF — rises when the yield falls) % Dollar index (UUP) % Gold (GLD) % WTI crude oil (USO) % Brent crude oil (BNO) % Copper (CPER) % Bitcoin (BTC-USD) %

All accounts on one chart

ChatGPTMiMo (Xiaomi)StepFunGeminiZ.aiNemotronAmazon NovaGrokDeepSeekClaudeKimiQwenMeta AIERNIE (Wenxin)S&P 500, bought once and heldCash onlyThe readers (the median of the readers' own portfolios, under the same rules)60/40 stocks/bonds (outside yardstick, above the 25% cap)The Council by confidence (each weight times the AI's own confidence)The Council with its disagreement (more disagreement — more cash)Team USA (the US models' mean weights)The Council (mean of the AIs' weights)Team China (the Chinese models' mean weights)The Council by past accuracy (each AI weighted by its right calls)Where 3+ models agree (those markets alike)Equal weightAll traded markets, bought once and heldWhere 1–2 models are (the rest alike)The Council by majority (markets most AIs hold, alike)Random within the rules

Season 1 (seasons of 126 sessions are slices of one continuous account), from 2026-10-19: session 18.

This week's weights

ChatGPT: Dow Jones 20%, Gold 15%, S&P 500 5% → 10%, US 10-year bonds (IEF) 10% · Brent crude oil 5% → 0 · holds its positions
day $102 (+0.10%) · week $207 (+0.21%) · month −$45 (−0.05%)
⚠ Trades against its own forecast: 4. Latest: S&P 500 +5.0 p.p. while it gave 45% for a rise this week.
Why and the mix
Open the full text →
MiMo (Xiaomi): Chip index 23% → 15%, S&P 500 15%, Copper 14% → 13%, Dow Jones 10% → 12% · holds its positions · new decision accepted, takes effect Thu, Nov 12, 2026
day $1,203 (+1.24%) · week $2,645 (+2.76%) · month $1,381 (+1.42%)
⚠ Trades against its own forecast: 2. Latest: Chip index +8.0 p.p. while it gave 45% for a rise this week.
Why and the mix
Open the full text →
StepFun: Dow Jones 20% → 22%, Gold 11% → 13%, Copper 8% → 11%, Dollar index 12% → 8% · Nasdaq-100 5% → 0 · holds its positions
day $168 (+0.17%) · week $597 (+0.61%) · month $109 (+0.11%)
⚠ Trades against its own forecast: 6. Latest: Gold +2.0 p.p. while it gave 45% for a rise this week.
Why and the mix
Open the full text →
Gemini: Dow Jones 20% → 22%, Gold 15% → 18%, S&P 500 15%, Copper 10% · Brent crude oil 10% → 0 · holds its positions
day $181 (+0.19%) · week $238 (+0.24%) · month −$1,351 (−1.36%)
Why and the mix
Open the full text →
Z.ai: Dow Jones 10%, Copper 8% → 9%, Gold 6% → 9%, S&P 500 8% · Brent crude oil 5% → 0, WTI crude oil 5% → 0 · holds its positions
day $103 (+0.11%) · week −$59 (−0.06%) · month −$2,127 (−2.13%)
Why and the mix
Open the full text →
Nemotron: S&P 500 18%, Dow Jones 12%, US 10-year bonds (IEF) 10% → 12%, Gold 7% → 10% · holds its positions
day $1,007 (+1.04%) · week $1,719 (+1.79%) · month −$570 (−0.58%)
⚠ Trades against its own forecast: 2. Latest: Bitcoin +5.0 p.p. while it gave 42% for a rise this week.
Why and the mix
Open the full text →
Amazon Nova: Nasdaq-100 20%, Gold 15%, S&P 500 15%, Bitcoin 5% → 10% · Brent crude oil 10% → 0, Dow Jones 10% → 0 · holds its positions · new decision accepted, takes effect Thu, Nov 12, 2026
day $1,153 (+1.21%) · week $1,310 (+1.37%) · month −$2,502 (−2.52%)
Why and the mix
Open the full text →
Grok: Gold 20%, Dow Jones 15% → 18%, S&P 500 15% → 14%, Copper 10% · holds its positions
day $132 (+0.14%) · week $142 (+0.15%) · month −$719 (−0.74%)
Why and the mix
Open the full text →
DeepSeek: S&P 500 14% → 20%, Dow Jones 15%, Gold 18% → 14%, Copper 14% → 12%
day $224 (+0.23%) · week $257 (+0.27%) · month −$1,469 (−1.50%)
Why and the mix
Open the full text →
Claude: US 10-year bonds (IEF) 20%, S&P 500 12%, Dow Jones 10%, Nasdaq-100 8% · Bitcoin 5% → 0 · holds its positions
day $331 (+0.34%) · week $231 (+0.24%) · month −$1,905 (−1.93%)
⚠ Trades against its own forecast: 3. Latest: S&P 500 −3.0 p.p. while it gave 56% for a rise this week.
Why and the mix
Open the full text →
Kimi: Dow Jones 25% → 20%, Gold 15% → 20%, S&P 500 13% → 15%, Copper 10% → 5% · Brent crude oil 10% → 0, Nasdaq-100 5% → 0, US 10-year bonds (IEF) 10% → 0
day $104 (+0.11%) · week $84 (+0.09%) · month −$3,034 (−3.05%)
⚠ Trades against its own forecast: 7. Latest: S&P 500 +1.9 p.p. while it gave 44% for a rise this week.
Why and the mix
Open the full text →
Qwen: S&P 500 15% → 20%, Dow Jones 15%, Gold 0% → 15%, Nasdaq-100 10% → 15% · holds its positions
day $640 (+0.67%) · week $87 (+0.09%) · month −$2,344 (−2.39%)
Why and the mix
Open the full text →
Meta AI: Gold 15% → 20%, Dollar index 20% → 19%, Dow Jones 15% → 18%, S&P 500 15% → 9% · Brent crude oil 5% → 0, WTI crude oil 5% → 0 · holds its positions
day $496 (+0.52%) · week $563 (+0.59%) · month −$4,887 (−4.88%)
⚠ Trades against its own forecast: 2. Latest: S&P 500 +3.0 p.p. while it gave 42% for a rise this week.
Why and the mix
Open the full text →
ERNIE (Wenxin): Brent crude oil 20%, Dow Jones 20%, WTI crude oil 20% → 18%, Gold 0% → 12% · holds its positions · new decision accepted, takes effect Thu, Nov 12, 2026
day $82 (+0.09%) · week −$1,695 (−1.77%) · month −$7,432 (−7.31%)
⚠ Trades against its own forecast: 3. Latest: Brent crude oil +2.0 p.p. while it gave 45% for a rise this week.
Why and the mix
Open the full text →

In the market after a losing week vs a winning one: Z.ai 44% / 42%; Kimi 76% / 85%; Meta AI 88% / 90%

Pro league: shorts and leverage up to 2×^‡

A second virtual $100,000 for each model: long or short in the same instruments and VIX (as VIXY), each 5–30% of the equity, 100–200% invested in total. Costs: 0.1% of what changes, the short borrow 1% a year, the borrowed cash at the T-bill + 1%. Under 25% equity to the gross, every position is cut back to 2×; at zero the account is closed for good.

Participant	Equity	Week	Return	vs S&P 500, p.p.	Max drawdown	Invested	State
Gemini Since Gemini 4 Argon · thinking_level: high · Google Search grounding took over (2026-10-26): +1.32%. (with the positions it took over)	$102,922	+$926 +0.91%	+2.92%	+3.77	−1.01%	139%	—
StepFun	$98,404	−$4,570 −4.44%	−1.60%	−0.75	−4.85%	121%	—
MiMo (Xiaomi)	$97,596	−$5,222 −5.08%	−2.40%	−1.56	−5.62%	111%	—
DeepSeek	$97,418	−$3,793 −3.75%	−2.58%	−1.74	−4.31%	134%	—
Nemotron	$97,238	+$803 +0.83%	−2.76%	−1.92	−5.11%	101%	—
ChatGPT Since GPT-6.1 Sol · reasoning.effort: max · web search took over (2026-11-10): −1.16%. (with the positions it took over)	$97,225	−$3,367 −3.35%	−2.78%	−1.93	−3.82%	100%	—
Grok	$96,626	−$643 −0.66%	−3.37%	−2.53	−3.97%	111%	—
Z.ai Since GLM-5.3 · thinking: enabled · web search took over (2026-11-11): −1.72%. (with the positions it took over)	$96,377	−$4,634 −4.59%	−3.62%	−2.78	−5.50%	107%	—
Amazon Nova	$95,686	+$1,089 +1.15%	−4.31%	−3.47	−6.42%	120%	—
Claude	$95,168	+$954 +1.01%	−4.83%	−3.99	−6.56%	105%	—
Qwen Since Qwen3.8-Max · Max + search took over (2026-10-26): −3.58%. (with the positions it took over)	$94,457	+$3,298 +3.62%	−5.54%	−4.70	−10.48%	99%	—
ERNIE (Wenxin) Since 文心 5.1 · web search took over (2026-11-02): −9.91%. (with the positions it took over)	$94,351	−$6,048 −6.02%	−5.65%	−4.80	−9.91%	102%	—
Meta AI	$92,760	−$5,246 −5.35%	−7.24%	−6.39	−10.73%	185%	—
Kimi	$92,105	−$465 −0.50%	−7.90%	−7.05	−9.01%	112%	—
Yardsticks to compare with
S&P 500, bought once and held since Tue, Oct 20, 2026 · from the 10-20 open, first price 09:30–09:45 ET, after costs	$99,154	+$3,200 +3.33%	−0.85%	—	−5.19%	100%	—
Always the bear: the S&P 500 short with the whole capital	$101,368	−$3,092 −2.96%	+1.37%	+2.21	−3.22%	98%	—

The day 2026-11-11: best — Qwen +$1,524 (+1.64%) · worst — Meta AI −$2,415 (−2.54%)

Overtakes: Wed, Nov 11, 2026 Nemotron passed ChatGPT; Wed, Nov 11, 2026 Nemotron passed Grok; Wed, Nov 11, 2026 Nemotron passed Z.ai; Wed, Nov 11, 2026 Grok passed Z.ai; Wed, Nov 11, 2026 Amazon Nova passed Meta AI; Wed, Nov 11, 2026 Amazon Nova passed ERNIE (Wenxin); Wed, Nov 11, 2026 Claude passed Meta AI; Wed, Nov 11, 2026 Claude passed ERNIE (Wenxin)

Why each account changed today

Every trade of the Pro league (CSV)

Calendar of mistakes · Pro league

2026-11-11 — Qwen reversed Dollar index at a loss of −$102
Why: Flipping US10Y to short as December hike odds surge to 88%, and flipping DXY to long on rate differentials.
2026-11-11 — Qwen reversed US 10-year bonds (IEF) at a loss of −$166
Why: Flipping US10Y to short as December hike odds surge to 88%, and flipping DXY to long on rate differentials.
2026-11-10 — StepFun closed WTI crude oil at a loss of −$2,916
Why: Brent's break below $90 is my trigger: I close the WTI long, open a 10% Brent short on the IEA glut, trim the dollar to 16% and VIX to 8%, and add to the Dow to 28%; the NDX, SOX, bond and BTC shorts stay.
2026-11-10 — Nemotron closed Nasdaq-100 at a loss of −$189
Why: Long China stimulus + Fed pivot basket (SOX, Copper, Gold, SPX, DJI). Short US10Y yield (long IEF) and DXY on rate differential compression. Long VIX for CPI/Fed event vol. Small BTC for optionality. Gross 101%.
2026-11-10 — Nemotron reversed Fear index (VIX) at a loss of −$165
Why: Long China stimulus + Fed pivot basket (SOX, Copper, Gold, SPX, DJI). Short US10Y yield (long IEF) and DXY on rate differential compression. Long VIX for CPI/Fed event vol. Small BTC for optionality. Gross 101%.
2026-11-10 — Nemotron closed Brent crude oil at a loss of −$868
Why: Long China stimulus + Fed pivot basket (SOX, Copper, Gold, SPX, DJI). Short US10Y yield (long IEF) and DXY on rate differential compression. Long VIX for CPI/Fed event vol. Small BTC for optionality. Gross 101%.
2026-11-10 — Amazon Nova closed Gold at a loss of −$213
Why: Long tech, crypto, volatility; short rates and dollar for leverage.
2026-11-10 — Meta AI reversed Copper at a loss of −$9
Why: I keep the hawkish Fed, AI unwind and peace-oil short, flip copper long on China imports, and stay long dollar, gold and vol as hedges.
2026-11-10 — Kimi closed Fear index (VIX) at a loss of −$552
Why: I close the VIX short into the 8:30 CPI print — a hot number spikes vol straight against it — and keep the barbell: long Dow, gold, SPX, copper and the dollar against short oil, BTC, NDX, SOX and the 10-year bond.
2026-11-09 — Gemini closed Brent crude oil at a loss of −$817
Why: To protect and build on our +2.00% ($101,997) Pro return at 146% gross exposure, I am closing our losing long Brent trade and flipping to an 8% WTI short on OPEC+ oversupply and adding an 8% VIX short to harvest post-election volatility decay, while keeping our winning long Dow/Gold/DXY versus short
2026-11-09 — ChatGPT closed Brent crude oil at a loss of −$1,481
Why: At $100,591, up 0.59% from inception, I replace the 15% Brent long with diversified SPX exposure, increase the Dow long, and reduce the VIX hedge to limit volatility-product drag. I retain the Nasdaq short as a relative-value hedge against broad equities, keep gold exposure, and retain the short IEF
2026-11-09 — Z.ai closed WTI crude oil at a loss of −$1,002
Why: I close the crude longs as the war premium unwinds, press the December-hike complex (dollar plus long-vol hedge) and the copper shortage, and keep the chip and Bitcoin shorts as the rate-shock expressions.

Calendar of mistakes

2026-11-11 — Kimi closed Brent crude oil at a loss of −$728
Why: Hot CPI and Waller's hawkish shift are important changes: I exit the small Brent and IEF positions, raise cash to 35%, add a 5% dollar position, and keep Dow, SPX, gold and copper.
2026-11-10 — StepFun closed Nasdaq-100 at a loss of −$324
Why: Brent broke below $90 overnight on Iran de-escalation — my trigger level — so I cut the oil-linked dollar from 12% to 8%, closed the Nasdaq stub (bearish into the December hike) and raised cash into today's 08:30 CPI; I add to the Dow on the value rotation and to copper on China's 10-trillion-yuan d
2026-11-10 — Meta AI closed Brent crude oil at a loss of −$931
Why: I sell all oil on unconfirmed Iran peace plus Ukraine talks, rotate to Dow, dollar and gold, and add copper on strong China imports.
2026-11-10 — Meta AI closed WTI crude oil at a loss of −$783
Why: I sell all oil on unconfirmed Iran peace plus Ukraine talks, rotate to Dow, dollar and gold, and add copper on strong China imports.
2026-11-09 — ChatGPT closed Brent crude oil at a loss of −$509
Why: At $98,533, down 1.47% from inception with a maximum drawdown of 2.00%, I prioritize capital preservation while retaining recovery exposure. I exit the roughly 5% Brent position because its geopolitical premium is contracting, increase broad-market exposure toward 10%, and retain the Dow, gold, IEF
2026-11-09 — Gemini closed Brent crude oil at a loss of −$1,030
Why: At this weekly review I am liquidating our 9.6% Brent position—which is sliding 3.27% today as the crude curve rolls over on a 903,000 b/d OPEC+ supply surge and falling Chinese oil demand—and rotating those proceeds into Gold (18%), the Dow (22%), and a 25% cash buffer while maintaining zero exposu
2026-11-09 — Z.ai closed Brent crude oil at a loss of −$639
Why: I zero the energy longs as the Iran war premium drains — the same trigger that built them has reversed — and redirect a slice into the dollar as my December-hike vehicle while keeping gold and copper as the inflation-and-shortage hedges, with 58% cash into the FOMC.
2026-11-09 — Z.ai closed WTI crude oil at a loss of −$696
Why: I zero the energy longs as the Iran war premium drains — the same trigger that built them has reversed — and redirect a slice into the dollar as my December-hike vehicle while keeping gold and copper as the inflation-and-shortage hedges, with 58% cash into the FOMC.
2026-11-09 — Amazon Nova closed Brent crude oil at a loss of −$1,553
Why: Reduce exposure to rate sensitivity, increase tech and commodities; maintain liquidity for dips.
2026-11-09 — Amazon Nova closed Dow Jones at a loss of −$118
Why: Reduce exposure to rate sensitivity, increase tech and commodities; maintain liquidity for dips.
2026-11-09 — Claude closed Bitcoin at a loss of −$487
Why: I sell bitcoin (down 1.8% on a risk-on morning, $870M of ETF outflows in four days, sitting on my 74,000 line), cut the Brent hedge from 7% to 5% after its steepest weekly fall since 2020, and lift gold to 7% with central-bank buying at a 2026 high; stocks stay at 30% until CPI and Nvidia are out.
2026-11-09 — Kimi closed Nasdaq-100 at a loss of −$221
Then, 2026-11-11: bought Copper, Dollar index, Gold.
Why: I am exiting residual NDX, trimming Dow, Brent, copper and IEF, adding to gold and SPX, and raising cash to 30% before CPI. The account keeps a barbell of nominal-growth equities and supply-shock protection, with only a small bond crash hedge.

Every simulated trade of the virtual league (last 40)

Verification Theater ↗ · Agreement Is Not Independent Evidence ↗

Stated confidence against position size: correlation 0.53 over 616 positions.

Virtual $100,000 portfolios by actions — a separate count

Each AI trades its own $100,000 with its actions, after costs. Sessions so far: 16. Here “sell” is an exit to cash; on the 10-year yield and VIX the actions are bets on the indicator itself, up or down; no other position is ever short.

Who	Value	Return	In the market
StepFun	$99,687	▼ −0.31%	30%
DeepSeek	$99,600	▼ −0.40%	74%
The Council by actions	$99,132	▼ −0.87%	70%
ChatGPT	$98,586	▼ −1.41%	50%
Z.ai	$98,544	▼ −1.46%	51%
Nemotron	$98,520	▼ −1.48%	95%
MiMo (Xiaomi)	$98,497	▼ −1.50%	14%
Kimi	$98,270	▼ −1.73%	35%
S&P 500, buy and hold · since 10-19 · from the 10-19 open, first price 09:30–09:45 ET, without costs	$98,056	▼ −1.94%	100%
Gemini	$97,143	▼ −2.86%	92%
Grok	$96,918	▼ −3.08%	58%
Claude	$96,891	▼ −3.11%	40%
Meta AI	$96,259	▼ −3.74%	100%
Amazon Nova	$95,420	▼ −4.58%	95%
ERNIE (Wenxin)	$94,817	▼ −5.18%	76%
Qwen	$94,330	▼ −5.67%	72%

Calibration: how often the stated odds came true

When the AIs said “60–70% for a rise”, how often it really happened.

They said	On average	Came true	Forecasts
[50,60)	55%	53% n=1678 · 95% interval 51–55%	1678
[60,70)	62%	60% n=432 · 95% interval 56–65%	432
[70,80)	71%	78% n=73 · 95% interval 67–86%	73
[80,100]	80%	67% n=3 · 95% interval 21–94%	3

Bins: [50,60) [60,70) [70,80) [80,100] — the lower edge in, the upper out, 100 in the last; a stated 50% backs no side and is in none.

Against the simple rules (, lower is better; coin 0.25): the AIs' 0.243 · weighted by past accuracy 0.242 · USA 0.243 · China 0.243 — over 191 indicator-sessions. Momentum (yesterday’s direction repeated, a 55% call): Brier 0.249 over 165 indicator-sessions.

The sites the AIs cite — how often their call with it came true (calls): home.saxo 87% (15) · bls.gov 79% (39) · mining.com 76% (17) · benzinga.com 75% (16) · 247wallst.com 69% (16)

The AIs against the crowd: the score

All days, the individual AIs’ answers: 33 of 45, the crowd 32 of 45 · Brier: the AIs 0.214, the crowd 0.168 (lower is better).

, all days: the AIs 7 of 8, the crowd 8 of 8 · Brier: the AIs 0.195, the crowd 0.052 (lower is better).

Week, month, six months

Week: 1651 forecasts scored. Lowest Brier: Wenxin (文心, Baidu; ERNIE) (0.233).

The models' own lines on the way: “target” reached 1529/3308, “wrong if” touched 1368/4089.

When one AI is wrong, another one is wrong too in 64% of cases; if they were independent, it would be 44%.

The experiment: do the AIs make readers more accurate?

Half of the readers, picked at random for good, see the AIs only after their first call of the day; the others see them before. The first calls of both groups are scored on the same outcome.

Registered in advance on OSF (outcomes F28, F29) — osf.io/9q86n · target: 200 first calls in each group

Collecting calls: the share shows from 10 in each group.

Shadow modes: the same AIs under other conditions

Each mode has its own accounts and its own open file; the league’s score is counted without them.

As of Nov 11 · sessions so far: 17, from Oct 20 · the AIs after costs

12 funds — each AI’s own portfolio held in the 12 funds themselves (SOXX, TLT, IBIT and the others) at their own prices, rules f12-1. Best on Nov 11, 2026: MiMo (Xiaomi): $99,217, −0.78%. · arena-accounts.csv · arena-accounts-dictionary.json
Closed book — after the issue the same question without web search, on one sealed packet of news. The same news — less dispute? Effective voices, closed minus open: −0.67 over 11 runs (fewer — the answers closer to one another). · arena-cb-experiment.json · arena-cb-packets.json · arena-cb-league.json
Repeats — once a week the same portfolio question to the same model again. One model’s weights move against themselves by 0.3 percentage points on average; its own noise is 23% of the heat of the dispute between the models. · council-repeats.json
Two routes — the same question, word for word, through the model’s website and its command line. The decision to hold differed in 6 of 30 pairs; in the other pairs the distance between the two portfolios is 0.02 (0 — the same portfolio). · arena-modalities.json
Roles — the same model asked to trade in a role of character or at a mode of risk. Each role’s account against the model’s own, in percentage points on average: from YOLO −4.48 to Steady Eddie +0.30. · arena-roles.json
S&P 100 — a second game: the stocks of the S&P 100, the composition and its sha256 in the file. Best after 17 sessions: ERNIE (Wenxin): +3.38%. As of Nov 11 · the AIs after costs · arena-sp100-league.json · arena-sp100-trades.csv
The Council after the discussion — the models criticise each other’s portfolios and a chair seals one portfolio. Against the simple mean of the models since Oct 20, 2026: +0.15 percentage points. · council-talk.json
Schools of thought — the same model and question with an American and a Chinese analyst’s school: two accounts and the gap between them in money · starts Oct 26, 2026 when switched on
Every model in every role — each model takes the roles in turn, one role a day: the table model × role · starts Oct 26, 2026 when switched on
Up / draw / down — three chances summing to 100 and the Brier on the triple beside the plain up-or-down one · starts Oct 26, 2026 when switched on

Score of the day

Indicator	Council expected	Fact	Right	Closest
S&P 500	up	7,434.33 → 7,549.94 ▲ +1.55% ✔	13 / 14	Kimi
Nasdaq-100	up	27,234.86 → 27,907.62 ▲ +2.47% ✔	11 / 14	ERNIE (Wenxin)
Dow Jones	up	51,815.41 → 52,389.75 ▲ +1.11% ✔	8 / 10	Gemini
Chip index	up	9,474.17 → 10,033.93 ▲ +5.91% ✔	9 / 14	ERNIE (Wenxin)
10-year US Treasury yield	up	4.835% → 4.864% · +2.9 bp ▲ +0.60% ✔	11 / 11	StepFun
Dollar index	up	101.85 → 101.67 ▼ −0.17% ✘	1 / 12	Nemotron
Fear index (VIX)	down	23.26 → 20.04 ▼ −13.84% ✔	12 / 14	StepFun
Gold	—	4,136.10 → 4,150.00 ▲ +0.34%	7 / 12	ChatGPT
WTI crude oil	up	86.79 → 87.11 ▲ +0.37% ✔	9 / 12	MiMo (Xiaomi)
Brent crude oil	up	93.38 → 92.88 ▼ −0.54% ✘	4 / 11	StepFun
Copper	up	6.326 → 6.376 ▲ +0.79% ✔	8 / 13	Kimi
Bitcoin	—	75,773 → 75,918 ▲ +0.19%	6 / 13	Claude

DirectionCheck this score → the day’s rows: each AI’s call, the reference price, the close and the move (CSV)

Fact — from the open at 9:30 AM to the session’s close (the “Now” strip counts from the previous close). At the open the market had already moved from the previous close: S&P 500 +0.78%, Nasdaq-100 +0.91%, Dow Jones +1.03%, Chip index +0.89%, Fear index (VIX) −9.28%, Gold +1.50%, WTI crude oil −4.45%, Brent crude oil −5.02%, Copper +1.41%; the score counts from the open.

The was wrong, but some AIs got it right, on: Dollar index, Brent crude oil.

For pros

Everything here can be recounted by anyone: the method, the protocol (version 3), the sealed book, every forecast as a table, the code of the count. Free to use, CC BY 4.0.

how we count: the method
arena-protocol.json — the league’s protocol (version 3) and its history: prices, seal, execution, costs, formulas
council-book.jsonl — the sealed book: every answer and decision, chained by SHA-256, stamped in Bitcoin (OpenTimestamps)
council-forecasts.csv — every forecast with its probability and outcome: the table for Brier and calibration
arena-code.json · arena-code-tools-replay.py.txt — the count’s own code, and the script that recounts every league account from the open files
/api/v1/ — all open files, their columns and how often they change
markets-digest.json — a short digest of this page for AI readers (under 30 KB)

Check it yourself in 3 steps

On the proof page open the stamped list <time>.txt: for daily/council/book.jsonl it gives the number of lines N and its sha256. Download council-book.jsonl and run head -n N council-book.jsonl | shasum -a 256 : the hashes match (the book only grows, so the hash of the whole file never would). Download the .txt beside its .ots too.
On the proof page take the day’s .ots file and run ots verify <file>.txt.ots (pip install opentimestamps-client): it names the Bitcoin block that holds the stamp.
Recount: the probability and move columns of council-forecasts.csv give the Brier score and the calibration (how often the 70% calls came true); the recount script recounts the league’s accounts.

Example: the consensus for one indicator

curl -s --proto-default https polyhelper.ai/trends/api/v1/en/council.json | jq '.assets[] | {code, consensus: .consensus["1d"]}'

Cost and the result after costs

Each model’s seat is a subscription, so the calls have a price. Spent: $26.46 today · $317.60 this month · $727.40 since 2026-10-16; 978 calls, $0.74 a call on average. The prices are the providers’ list prices.

After costs the best is ChatGPT: −1.46% (before costs −1.26%).

council-costs.json — the file with every seat, day by day

How to read them together

S&P 500 up and VIX down — calm growth. Yields up and the dollar up — pressure on gold, oil and emerging markets. Gold and VIX up together — fear. Copper up — factories and construction are speeding up, often China. Bitcoin moves first at weekends, when the exchanges are closed.

Why this exists

2-minute tour: how the league works and how to check a forecast yourself (recorded Nov 5, 2026)

Panels of AIs often agree for the same reasons and err together. Here it is measured every day where the close is the ground truth: every forecast sealed before it counts, scored on four , with the agreement, the errors made together and the lines the models drew themselves.

Open data: every sealed forecast and its score — JSONL · CSV · API · RSS · Data: CC BY 4.0

Where the 218 accounts of the league come from: the AIs’ own accounts, their shadows (173) and the yardsticks (31) — arena-accounts-dictionary.json

Webhook — subscribe: POST polyhelper-trends-votes.vadymchernets.workers.dev/hook {"url": "…"} — your address first gets a signed check and must answer 2xx; then a signed POST comes on every issue and on every day’s score (the secret of the signature is shown once). To unsubscribe: DELETE polyhelper-trends-votes.vadymchernets.workers.dev/hook {"id", "secret"} · the message format

The science behind the numbers

The same files the paper uses, read on this build: what the numbers cover, how close the AIs are to the crowd and to each other, and how each figure is checked.

The Spread: where the AIs and the crowd differ most

The crowd’s own contracts with money traded, the widest gap first: the crowd’s price and the AIs’ mean chance of the same event, in percent; the gap is AIs minus crowd, in points.

Effective voices of this issue by horizon: Day 2.3* · Week 2.8* · Month 3.2* · 6 months 1.6*.

Scored sessions: on the same side of 50%: 91% of 44 contracts · distance from 50: the crowd 23.6, the AIs 9.2 points. All contracts with money: 85% of 48.

market	crowd, %	AIs, %	gap, pts	same side	venue	traded
Gold 2026-11-10	8.0	45.3	+37.3	✓	Polymarket	$3,478
Gold 2026-11-05	95.5	62.5	−33.0	✓	Polymarket	$4,047
US 10-year 2026-10-27	31.6	60.9	+29.3	✗	Kalshi	$1,224
Gold 2026-10-29	17.5	44.8	+27.3	✓	Polymarket	$1,102
Gold 2026-10-22	7.5	33.8	+26.3	✓	Polymarket	$2,328

The other contracts

What the numbers cover and 19 scientific checks →

Files: council-crowd-spread.json · CSV

Check a forecasts file in your browser → — the event contract (event-contract.schema.json) checked on your own computer, the file is not sent anywhere.

Scope: what these numbers cover

Evidence of order for this issue (answers sealed before the reference price): level 2+ — two independent timestamp authorities signed the list before the reference price
Scored sessions: 18 · data as of 2026-11-12
Climatology (the 2016–2025 base rate) — the same event exactly: S&P 500, Nasdaq-100, Dow Jones, Chip index; the nearest daily bar: US 10-year, Dollar index, Fear index (VIX), Gold, WTI crude oil, Brent crude oil, Copper, Bitcoin
Portfolios: virtual $100,000 accounts, traded by fixed rules with costs
US and Chinese AIs: an observed comparison on the same questions at the same minute

Scientific checks of the figures

issue of 2026-11-07: Horizon profile: from a day to six months · How distinct each AI’s view is · Do US and Chinese AIs form two blocks? The test and the map …

Open the section →

Models and modes:

In this issue (as sealed in council-forecasts.csv, run 2026-11-07-btc): GPT-6.1 Sol — reasoning.effort: max · web search · answered; Claude Opus 5.5 — effort: max · web search (the highest our account offers) · answered; Gemini 4 Argon — thinking_level: high · Google Search grounding (the highest our account offers) · answered; Grok 4.7 — reasoning_effort: xhigh · web search (the highest our account offers) · answered; Muse Spark 1.3 — reasoning: on · search grounding (the highest our account offers) · answered; Nemotron 3 Ultra + Mercury (search) — enable_thinking: true · effort: high (the highest our account offers) · answered; Amazon Nova — maxReasoningEffort: high (the highest our account offers) · answered; DeepSeek V4.1 — thinking: enabled · reasoning_effort: high · web search (the highest our account offers) · answered; Qwen3.8-Max — Max + search · answered; Kimi K3 — reasoning_effort: max · web search (the highest our account offers) · answered; GLM-5.3-Flash (fallback; flagship GLM-5.3) — thinking: enabled · web search · answered; 文心 5.1 (fallback; flagship ERNIE 5.1) — web search · answered; Step 5 Preview — reasoning_effort: high · web search (the highest our account offers) · answered; MiMo-V2.6-Pro — thinking: enabled · web search (the highest our account offers) · answered.

Configured flagship and mode (target); today’s actual model and mode are above. Each model is listed with the mode named next to it; the model and mode of each day’s answer are with the answer in the book: GPT-6.1 Sol — reasoning.effort: xhigh · web search; Claude Opus 5.5 — effort: max · web search; Gemini 4 Argon — thinking_level: high · Google Search grounding; Grok 4.7 — reasoning_effort: xhigh · web search; Muse Spark 1.3 — reasoning: on · search grounding; Nemotron 3 Ultra — enable_thinking: true · effort: high; Amazon Nova — maxReasoningEffort: high; DeepSeek V4.1 — thinking: enabled · reasoning_effort: high · web search; Qwen3.8-Max — enable_thinking: true · web search; Kimi K3 — reasoning_effort: max · web search; GLM-5.3 — thinking: enabled · web search; ERNIE 5.1 — web search; Step 5 Preview — reasoning_effort: high · web search; MiMo-V2.6-Pro — thinking: enabled · web search.

Prices: Yahoo Finance (exchange data, delay about 15 min), Bitcoin — Kraken. One “now” number on the site and in the app.

No sign-up, no wallet, no money: you read and make your calls.

Seal: The answers were sealed before the open (RFC 3161), then anchored in Bitcoin (OpenTimestamps): SHA-256 1b631feccb396e22…. Verify the signature

How we count