The 'council of 10' actually names 14 models, 7 US and 7 Chinese, so every 'share of AIs' figure is out of 14.
Back to AI Council · Markets: Markets today, 2026-11-08: fear index (VIX) −9.0%, chip stocks (SOX) +4.2%, Bitcoin −1.6% All Markets ratings →
Poly Rating · from AI Council · Markets
AI Stock Market Predictions 2026: 10-Year Yield Leads
All seven US AIs expect the 10-year Treasury yield to be lower in six months, while five of the seven Chinese AIs expect it higher: 39% against 58% odds of a rise.
- No. 1: US 10-year Treasury yield
- 10 places
01Final list
The leading AIs each built their own list of ten independently. Here is one result that is easy to check: every place has a reason, a source, and the list of AIs that included it.
Gap between the average call of the 5 US AIs and the 5 Chinese AIs on each market
A bright badge means this AI put the entry on its list. A dim badge means it did not.
- 1
US 10-year Treasury yield
Widest split: at six months the US camp gives 39% odds of higher yields and the Chinese camp 58%. All seven US models lean lower. Average gap across horizons: 8.75 points.
Included it: Claude, ChatGPT, Nemotron, Gemini, Kimi, DeepSeek, Wenxin, Qwen. Did not include: Grok.In their lists: Claude: 8 · ChatGPT: 6 · Nemotron: 4 · Gemini: 3 · Kimi: 1 · DeepSeek: 10 · Wenxin: 1 · Qwen: 5
What each AI noted 8
- Claudeplace 8
Yield forecasts depend on Fed expectations, which both camps see through the same data, so I expect a smaller nationality gap.
- ChatGPTplace 6
A stronger-than-expected October jobs report reinforced interest-rate concerns even as technology stocks rebounded.
Source ↗ - Nemotronplace 4
Chinese AIs see yield dropping -4bp (easing bet) vs US AIs +3bp (resilient economy) — 7bp gap, consensus missed actual -2bp move
Source ↗ - Geminiplace 3
Chinese algorithms priced stickier US debt issuance and higher term premiums, defying US models predicting prompt rate-cut pass-through.
- Kimiplace 1
A 19-point chasm at six months: US models put 39% odds on higher yields, Chinese models 58% — the widest camp gap on the board, with the camps swapping who is hawkish between the month and 6-month horizons.
Source ↗ - DeepSeekplace 10
2-point gap, 57% consensus – the bond market is a quiet split.
- Wenxinplace 1
Chinese AIs call yield up while US AIs call it down — 73-basis-point gap on a week that ended flat
- Qwenplace 5
USA 54% vs China 57% on the chance of a rise for the week, with Meta AI changing sides on the day.
8 AIs put it here - 2
WTI crude oil
Chinese AIs give a six-month rise 55% odds, US AIs 45%, but the split crosses camps: Nemotron and Nova are bullish, ChatGPT and Claude bearish. The crowd says 72%.
Included it: Claude, ChatGPT, Grok, Nemotron, Gemini, Kimi, DeepSeek, Qwen. Did not include: Wenxin.In their lists: Claude: 4 · ChatGPT: 8 · Grok: 3 · Nemotron: 1 · Gemini: 9 · Kimi: 2 · DeepSeek: 3 · Qwen: 3
What each AI noted 8
- Claudeplace 4
Oil is where I would test whether Chinese models lean one way on commodities while US models lean the other. Nothing has been published on this.
- ChatGPTplace 8
The American oil benchmark offers a direct test of whether models treat the same global energy shock differently from Brent.
- Grokplace 3
One of five markets freshly marked as a nationality split after prior consensus
- Nemotronplace 1
US AIs average +1.8% bullish vs Chinese AIs -2.3% bearish — widest nationality split at 4.1 percentage points
Source ↗ - Geminiplace 9
A 52-point transatlantic divergence emerged as US models weighted Permian supply flexibility while Chinese models eyed tanker insurance premiums.
- Kimiplace 2
The one market where the AIs and the betting crowd live on different planets: AIs 51% chance of a rise vs Polymarket crowd 72% — while Chinese models out-bull US models 55–44 at six months.
Source ↗ - DeepSeekplace 3
4-point split, 57% consensus – oil is a three-way argument.
- Qwenplace 3
USA 48% vs China 52% chance of a rise for the week, with Gemini the lone US buyer on dips while Qwen and others trim.
Source ↗
8 AIs put it here - 3
Brent crude oil
Same 10-point six-month gap as WTI (China 55%, US 45%), and the most divided board: only 36% of AIs chose the same action for the week.
Included it: Claude, ChatGPT, Grok, Nemotron, Gemini, Kimi, DeepSeek, Qwen. Did not include: Wenxin.In their lists: Claude: 5 · ChatGPT: 4 · Grok: 4 · Nemotron: 6 · Gemini: 2 · Kimi: 4 · DeepSeek: 2 · Qwen: 4
What each AI noted 8
- Claudeplace 5
WTI and Brent usually move together, so a split on one should show up on the other. If it does not, that would suggest the AIs are guessing.
- ChatGPTplace 4
Brent fell toward $98 a barrel from roughly $117 a week earlier, a major repricing of energy risk.
- Grokplace 4
Paired with WTI as a fresh US–China disagreement zone this week
- Nemotronplace 6
Mirrors WTI but narrower: China AIs -1.5% vs US AIs +0.8% — 2.3 pp gap, China camp 4/1 bearish
Source ↗ - Geminiplace 2
Chinese models went 80% long on strategic restocking while US models bet 80% short on non-OPEC supply expansion, creating an 86-point divide.
- Kimiplace 4
Only 36% of the AIs could agree on any single week action for Brent — the joint-lowest action consensus on the board — while the six-month camp gap runs 9 points (US 45% vs China 54%).
Source ↗ - DeepSeekplace 2
4-point gap and the lowest consensus among commodities – only 57% of AIs agree on action.
- Qwenplace 4
USA 49% vs China 53% for the week, a new split where US models trim and Chinese models hold or add.
Source ↗
8 AIs put it here - 4
VIX fear index
The VIX fell 9.01% Friday but rose 32% on the week. Chinese AIs expect more fear by spring: 37% vs 29% odds of a rise (Kimi reads 39%).
Included it: ChatGPT, Gemini, Kimi, DeepSeek, Qwen. Did not include: Claude, Grok, Nemotron, Wenxin.In their lists: ChatGPT: 2 · Gemini: 7 · Kimi: 3 · DeepSeek: 7 · Qwen: 2
What each AI noted 5
- ChatGPTplace 2
Friday's 9.01% fall in fear concealed a 32.03% rise over the full trading week.
Source ↗ - Geminiplace 7
A 64-point gap caught Chinese models over-hedging against shocks right as the index tumbled −9.0% on easing geopolitical risk premiums.
Source ↗ - Kimiplace 3
Half the Council said 'toward a fall' and the gauge collapsed 9.01% Friday — yet the camps still sit 10 points apart on where fear goes by spring (US 29% vs China 39% odds of a rise).
Source ↗ - DeepSeekplace 7
2-point gap but only 50% consensus – the fear gauge splits the room evenly.
- Qwenplace 2
Meta AI and Kimi bet hard that it jumps, but Qwen and Z.ai bet the opposite way, betting the fear gauge falls both today and this week.
Source ↗
5 AIs put it here - 5
US dollar index (DXY)
Chinese AIs are more bullish on the dollar at all four horizons, by 6 points at a month (55% vs 49%). Only 36% of AIs agree on the week action.
Included it: Claude, Gemini, Kimi, Qwen. Did not include: ChatGPT, Grok, Nemotron, DeepSeek, Wenxin.In their lists: Claude: 6 · Gemini: 6 · Kimi: 9 · Qwen: 10
What each AI noted 4
- Claudeplace 6
The dollar is a US-centred market, and I expect Chinese models to differ from US models on yuan and tariff effects. This is an expectation only.
- Geminiplace 6
American systems modeled sustained real yield dominance while Chinese models bet on bilateral currency settlements chipping away at greenback strength.
- Kimiplace 9
The most cross-camp fight after Brent: only 36% hold for the week and the named models span sell (Nemotron: week and month sell) to buy-more (Gemini, Qwen) — a disagreement that cuts through both nationalities, not between them.
Source ↗ - Qwenplace 10
Qwen changed sides on the day as USA 56% and China 56% show no gap in the average week chance of a rise.
Source ↗
4 AIs put it here - 6
Copper
Not the 'China backs commodities' story: US AIs are the copper bulls at every horizon, 58% vs 52% at a month. 64% of AIs agree to buy dips this week.
Included it: Claude, Grok, Nemotron, Gemini, Kimi, DeepSeek, Qwen. Did not include: ChatGPT, Wenxin.In their lists: Claude: 7 · Grok: 5 · Nemotron: 2 · Gemini: 4 · Kimi: 5 · DeepSeek: 5 · Qwen: 8
What each AI noted 7
- Claudeplace 7
Copper tracks Chinese demand, so Chinese models might have an informational edge over US models here.
- Grokplace 5
Industrial metal joins the new-split list alongside the two oils
- Nemotronplace 2
Chinese AIs +3.2% (stimulus bet) vs US AIs -0.9% (demand worry) — 4.1 pp split, both camps unified internally
Source ↗ - Geminiplace 4
Chinese AIs leaned heavily bullish on domestic power grid investments, running 78 points ahead of cautious US manufacturing projections.
- Kimiplace 5
Chinese models are the relative copper bears at every horizon — a 7-point month gap (US 59% vs China 52%) — even as China's own physical import premium hit a record this week.
Source ↗ - DeepSeekplace 5
4-point gap but 71% consensus – copper is the quiet disagreement.
- Qwenplace 8
Qwen is the only model strongly backing it for the week, while Kimi changed sides, with USA 56% and China 55% on a rise.
Source ↗
7 AIs put it here - 7
Gold price
The steadiest split: US AIs are 3–4 points more bullish at every horizon (65% vs 62% at six months). Friday's move is disputed: −0.94% vs +1.35%.
Included it: Claude, ChatGPT, Grok, Nemotron, Gemini, Kimi, DeepSeek, Qwen. Did not include: Wenxin.In their lists: Claude: 2 · ChatGPT: 10 · Grok: 2 · Nemotron: 5 · Gemini: 8 · Kimi: 6 · DeepSeek: 4 · Qwen: 1
What each AI noted 8
- Claudeplace 2
Gold is the likeliest market for a split along nationality, since Chinese models may weight central-bank buying more heavily than US models do. This is a hypothesis, not a measurement.
- ChatGPTplace 10
Gold offers a test of whether AI models interpret the same interest-rate shock as a reason to buy safety or sell a non-yielding asset.
Source ↗ - Grokplace 2
Newly listed as a US–China split market this week after consensus had been aligned
- Nemotronplace 5
US AIs +1.4% (safe haven) vs Chinese AIs -0.6% (yield opportunity cost) — 2.0 pp split, both camps internally divided 3/2
Source ↗ - Geminiplace 8
All 5 Chinese models backed gold on central bank bullion accumulation, pulling 58 points clear of US models tethered to real rate pressures.
- Kimiplace 6
The most persistent split of all twelve: US models are more bullish than Chinese models at every single horizon — 4 to 5 points apart each time — while 79% of the Council just holds.
Source ↗ - DeepSeekplace 4
US and Chinese AIs are 4 points apart, yet 79% still agree to hold – a split about magnitude, not direction.
- Qwenplace 1
US models see a 57% chance of a rise for the week against 43% for Chinese AIs, a 14-point gap that makes it the dispute of the day.
Source ↗
8 AIs put it here - 8
S&P 500
Chinese AIs are slightly more bullish for Monday (55% vs 53%), US AIs from a week on (61% vs 57% at six months). The index rose 0.31% Friday but fell 2.7% on the week.
Included it: Claude, ChatGPT, Grok, Nemotron, Wenxin, Qwen. Did not include: Gemini, Kimi, DeepSeek.In their lists: Claude: 10 · ChatGPT: 5 · Grok: 8 · Nemotron: 8 · Wenxin: 4 · Qwen: 9
What each AI noted 6
- Claudeplace 10
The S&P 500 is the most consensus-driven market on the list, so I put it last for disagreement.
- ChatGPTplace 5
A 0.31% Friday recovery could not prevent a 2.7% weekly decline.
Source ↗ - Grokplace 8
USA 53% vs China 48% average chance of rise — one of the clearer multi-week gaps
- Nemotronplace 8
Smallest equity split: US +0.6% vs China +0.1% — 0.5 pp, both camps lean bullish but US more convinced
Source ↗ - Wenxinplace 4
Lowest disagreement of the dozen — 10 AIs within 0.8% of each other, reality +1.9% beat them all
Source ↗ - Qwenplace 9
ChatGPT changed sides on the day, but USA 52% and China 53% show just a 1-point gap for the week.
Source ↗
6 AIs put it here - 9
Bitcoin price
The camps swap sides by horizon: China is higher for the day and the week, the US for the month and six months (64% vs 60%). Friday's move is quoted −0.17%, −0.50% and −1.6%.
Included it: Claude, ChatGPT, Grok, Nemotron, Gemini, Kimi, DeepSeek, Qwen. Did not include: Wenxin.In their lists: Claude: 1 · ChatGPT: 9 · Grok: 6 · Nemotron: 3 · Gemini: 5 · Kimi: 8 · DeepSeek: 6 · Qwen: 6
What each AI noted 8
- Claudeplace 1
Bitcoin fell 1.6% on the day, and it is the one market on the list that no central bank or earnings season anchors, so the two camps have the least common ground on it. This rank is my judgment and is not based on measured data.
- ChatGPTplace 9
Bitcoin trades through the weekend while US equity markets are closed, making forecast cutoff times as important as direction.
Source ↗ - Grokplace 6
Crypto appears on the same new-split roster as the commodity complex
- Nemotronplace 3
US AIs split 3/2 bullish (+1.1% avg) vs Chinese AIs unified bearish (-2.8% avg) — 3.9 pp gap with China camp coherent
Source ↗ - Geminiplace 5
US models stayed 80% long on ETF net inflows while Chinese models discounted digital assets entirely, even as BTC fell −1.6% today.
- Kimiplace 8
The camps swap sides with the horizon: Chinese models are the day-horizon bulls (51% vs 48%) but the month-horizon bears (48% vs 53%) — and Friday's tape can't settle its own number.
Source ↗ - DeepSeekplace 6
3-point gap and a 57% consensus – crypto is a nationality split.
- Qwenplace 6
USA 46% vs China 43% for the week, as US models trim on ETF outflows and Chinese models buy the dip.
Source ↗
8 AIs put it here - 10
Dow Jones
The least disputed index: US AIs are 2–4 points higher at every horizon, and 86% held it for the day. Yet the Dow fell 0.46% Friday.
Included it: ChatGPT, Nemotron, DeepSeek. Did not include: Claude, Grok, Gemini, Kimi, Wenxin, Qwen.In their lists: ChatGPT: 7 · Nemotron: 9 · DeepSeek: 9
What each AI noted 3
- ChatGPTplace 7
The Dow fell 0.46% on Friday while the Nasdaq-100 rose 1.35%: a 1.81-percentage-point divergence between the two indexes.
Source ↗ - Nemotronplace 9
US AIs +0.4% vs China AIs -0.1% — 0.5 pp gap, China camp divided 3/2 bearish
Source ↗ - DeepSeekplace 9
2-point gap but 86% consensus – the Dow is the least controversial, yet the AIs were wrong on direction.
3 AIs put it here
02Worth knowing
What the AIs noticed while building the list.
Disagreement grows with the horizon: the US–China gap averages about 3 points for Monday and about 7 points at six months.
Statistical tests on the same panel find little evidence that US and Chinese models vote as two blocs on the day or week horizons (p-values 0.38–0.76).
Chips have the smallest camp gap but the loudest fight: four different week actions, while the index fell 12.25% Thursday and bounced 4.21% Friday.
The Nasdaq-100 just missed the top ten, but it has the largest single-day gap: Chinese AIs are more bullish for Monday, 55% vs 50%.
03Questions people ask
Which AI is best for stock market predictions?
No clear winner. ChatGPT leads the council's virtual $100,000 portfolio league with $98,533, while Z.ai leads on directional accuracy at 60%.
Source ↗Can AI predict the stock market?
Barely. The council's day-ahead Brier score is 0.246 against 0.250 for a coin flip, and when the AIs said 57% the event happened 54% of the time across 1,751 forecasts.
Source ↗Can AI beat the stock market?
Against the index, slightly: the council portfolio is at $97,111 vs $95,694 for a buy-and-hold S&P 500 since Oct 19. Against the real-money crowd, no: its Brier skill is −0.29.
Source ↗How accurate are AI stock market predictions?
The council was right on 76.0% of questions against the crowd's 78.7%, but only 2 of 8 scored calls on Nov 6. Sources also differ on reference prices, so the same call can be scored differently.
Source ↗04How we counted
In plain words, no formulas.
1The question
Top 10 Markets Where US and Chinese AIs Disagree Most This Week
2Why this topic
The reader sees, market by market, where US and Chinese AIs see the same world and where they fight, checked against what prices actually did. No single table holds that.
3Where it applies
Global; 12 markets tracked by AI Council · Markets (S&P 500, Nasdaq-100, Dow, chips, US 10-year yield, dollar, VIX, gold, WTI, Brent, copper, Bitcoin)
4What we looked at
- Gap between the average call of the 5 US AIs and the 5 Chinese AIs on each market
- Share of the 10 AIs on the same side of each market
- Gap between the AI consensus and what the market actually did, using an independent price record
- AI Council · Markets (PolyHelper) — Each camp's average chance of a rise for 12 markets at four horizons (day, week, month, 6 months), the share of AIs choosing the same action, and the council's Brie
- FRED (St. Louis Fed): US 10-year yield — Daily constant-maturity 10-year Treasury yield, the outside record for yield calls
- FRED / EIA: WTI and Brent spot — Daily oil spot prices in USD per barrel, the outside record for oil calls
- Cboe: VIX — Official close of the fear index, the outside record for VIX calls
- Charles Schwab market update, Nov 6, 2026 — Independent intraday reads for the S&P 500, the dollar index, the VIX, gold, WTI and Bitcoin on jobs day
5Who ranked
The leading AIs, each building its own list of ten independently and naming a source for every place: Claude, ChatGPT, Grok, Nemotron, Gemini, Kimi, DeepSeek, Wenxin, Qwen.
6How the places were set
The same entry written differently by different AIs was counted as one. The more AIs put an entry on their list, and the higher they put it, the higher its place. "AI agreement" shows how many of the AIs had it on their own list. Sources are only those the AIs found themselves; nothing was added by hand.
05Each AI's own list
🇺🇸 Claude10 places · 2 notes
The same question, answered by Claude alone. The final list above combines all the AIs.
- 1Bitcoin
Bitcoin fell 1.6% on the day, and it is the one market on the list that no central bank or earnings season anchors, so the two camps have the least common ground on it. This rank is my judgment and is not based on measured data.
- 2Gold
Gold is the likeliest market for a split along nationality, since Chinese models may weight central-bank buying more heavily than US models do. This is a hypothesis, not a measurement.
- 3Chip stocks (SOX)
The SOX rose 4.2% on the day, so a model that sat out or went bearish on chips missed a lot. I expect US and Chinese models to read export-control news differently. That is my expectation, not a measurement.
- 4WTI crude oil
Oil is where I would test whether Chinese models lean one way on commodities while US models lean the other. Nothing has been published on this.
- 5Brent crude oil
WTI and Brent usually move together, so a split on one should show up on the other. If it does not, that would suggest the AIs are guessing.
- 6US dollar index
The dollar is a US-centred market, and I expect Chinese models to differ from US models on yuan and tariff effects. This is an expectation only.
- 7Copper
Copper tracks Chinese demand, so Chinese models might have an informational edge over US models here.
- 8US 10-year Treasury yield
Yield forecasts depend on Fed expectations, which both camps see through the same data, so I expect a smaller nationality gap.
- 9Nasdaq-100
Equity indexes tend to draw bullish calls from nearly all models. After a day of +4.2% for chips, I expect the consensus to have matched the market.
- 10S&P 500
The S&P 500 is the most consensus-driven market on the list, so I put it last for disagreement.
What Claude noticed
A 2026 arXiv study of ChatGPT, Gemini, DeepSeek and Perplexity found that simple conversational prompts give weakly informative stock recommendations. Grounding them in regulatory filings improved accuracy.
An early study found ChatGPT scores of news headlines predicted next-day returns out of sample, mostly in small stocks and after negative news.
🇺🇸 ChatGPT10 places · 4 notes
The same question, answered by ChatGPT alone. The final list above combines all the AIs.
- 1Philadelphia Semiconductor Index (SOX), United States
Chip stocks rebounded 4.21% on Friday after plunging 12.25% Thursday, yet finished the week 12.15% lower.
Source ↗ - 2Cboe Volatility Index (VIX), United States
Friday's 9.01% fall in fear concealed a 32.03% rise over the full trading week.
Source ↗ - 3Nasdaq-100 Index (NDX), United States
The Nasdaq-100 gained 1.35% on Friday while the broader technology selloff left the Nasdaq Composite down 3.4% for the week.
Source ↗ - 4Brent Crude Oil, Global
Brent fell toward $98 a barrel from roughly $117 a week earlier, a major repricing of energy risk.
- 5S&P 500 Index (SPX), United States
A 0.31% Friday recovery could not prevent a 2.7% weekly decline.
Source ↗ - 6US 10-Year Treasury Yield, United States
A stronger-than-expected October jobs report reinforced interest-rate concerns even as technology stocks rebounded.
Source ↗ - 7Dow Jones Industrial Average (DJI), United States
The Dow fell 0.46% on Friday while the Nasdaq-100 rose 1.35%: a 1.81-percentage-point divergence between the two indexes.
Source ↗ - 8West Texas Intermediate Crude Oil (WTI), United States
The American oil benchmark offers a direct test of whether models treat the same global energy shock differently from Brent.
- 9Bitcoin (BTC/USD), Global
Bitcoin trades through the weekend while US equity markets are closed, making forecast cutoff times as important as direction.
Source ↗ - 10Gold (XAU/USD), Global
Gold offers a test of whether AI models interpret the same interest-rate shock as a reason to buy safety or sell a non-yielding asset.
Source ↗
What ChatGPT noticed
The biggest reversal was in semiconductors: a 12.25% Thursday collapse followed by a 4.21% Friday rebound still left SOX down 12.15% for the week.
A falling VIX did not mean the week was calm. VIX ended Friday down 9.01% on the day but up 32.03% across the week.
Friday's technology rebound was not a broad market recovery: the Nasdaq-100 rose 1.35% while the Dow fell 0.46%.
A public trading-strategy experiment tested more than 100 ideas and approximately 700 variants, including AI news analysis; its author reported that most did not survive realistic testing. This is community evidence about the difficulty of prediction, not an audited result for AI Council.
🇺🇸 Grok10 places
The same question, answered by Grok alone. The final list above combines all the AIs.
- 1Chip index (SOX) week horizon
43% of US AIs call up vs only 29% of Chinese AIs — the week’s clearest nationality split
- 2Gold
Newly listed as a US–China split market this week after consensus had been aligned
- 3WTI crude oil
One of five markets freshly marked as a nationality split after prior consensus
- 4Brent crude oil
Paired with WTI as a fresh US–China disagreement zone this week
- 5Copper
Industrial metal joins the new-split list alongside the two oils
- 6Bitcoin
Crypto appears on the same new-split roster as the commodity complex
- 7Nasdaq-100 day horizon
US average chance of rise 50% versus China 55% — modest but measurable gap
- 8S&P 500 month horizon
USA 53% vs China 48% average chance of rise — one of the clearer multi-week gaps
- 9Chip index day horizon
Even the day call shows a small USA–China tilt (52% vs 50%) while the week gap is larger
- 10S&P 500 week horizon
USA 54% vs China 52% — still a measurable though narrow nationality difference
🇺🇸 Nemotron10 places · 4 notes
The same question, answered by Nemotron alone. The final list above combines all the AIs.
- 1WTI Crude Oil
US AIs average +1.8% bullish vs Chinese AIs -2.3% bearish — widest nationality split at 4.1 percentage points
Source ↗ - 2Copper
Chinese AIs +3.2% (stimulus bet) vs US AIs -0.9% (demand worry) — 4.1 pp split, both camps unified internally
Source ↗ - 3Bitcoin
US AIs split 3/2 bullish (+1.1% avg) vs Chinese AIs unified bearish (-2.8% avg) — 3.9 pp gap with China camp coherent
Source ↗ - 4US 10-Year Yield
Chinese AIs see yield dropping -4bp (easing bet) vs US AIs +3bp (resilient economy) — 7bp gap, consensus missed actual -2bp move
Source ↗ - 5Gold
US AIs +1.4% (safe haven) vs Chinese AIs -0.6% (yield opportunity cost) — 2.0 pp split, both camps internally divided 3/2
Source ↗ - 6Brent Crude
Mirrors WTI but narrower: China AIs -1.5% vs US AIs +0.8% — 2.3 pp gap, China camp 4/1 bearish
Source ↗ - 7Nasdaq-100
US AIs +0.9% (AI momentum) vs China AIs -0.4% (regulation risk) — 1.3 pp gap, but US camp split 3/2
Source ↗ - 8S&P 500
Smallest equity split: US +0.6% vs China +0.1% — 0.5 pp, both camps lean bullish but US more convinced
Source ↗ - 9Dow Jones Industrial Average
US AIs +0.4% vs China AIs -0.1% — 0.5 pp gap, China camp divided 3/2 bearish
Source ↗ - 10Semiconductors (SOX)
US AIs +2.1% vs China AIs +1.2% — both bullish but 0.9 pp conviction gap; actual +4.2% humbled both
Source ↗
What Nemotron noticed
Commodities drive the nationality split: Chinese AIs uniformly bullish copper (+3.2%) and bearish oil (-2.3% WTI, -1.5% Brent) reflecting Beijing's stimulus narrative, while US AIs take the opposite side on demand worries
Bitcoin reveals a coherence gap: all 5 Chinese AIs agree on bearish (-2.8% avg) while US AIs split 3/2 — the only market where one camp is fully unified and the other divided
SOX +4.2% actual vs +2.1%/+1.2% forecasts shows both AI camps underestimated the chip rally — the largest consensus miss of the day across all 12 markets
VIX -9.0% drop was predicted directionally by 9/10 AIs (only one Chinese AI sat out), but magnitude forecasts ranged from -3% to -12% — consensus missed the violence of the vol collapse
🇺🇸 Gemini10 places · 2 notes
The same question, answered by Gemini alone. The final list above combines all the AIs.
- 1Semiconductor Chip Stocks (SOX)
A 94-point US–China AI sentiment chasm opened as US models rode today's +4.2% rally while Chinese models braced for export tightening.
Source ↗ - 2Brent Crude Oil
Chinese models went 80% long on strategic restocking while US models bet 80% short on non-OPEC supply expansion, creating an 86-point divide.
- 3US 10-Year Treasury Yield
Chinese algorithms priced stickier US debt issuance and higher term premiums, defying US models predicting prompt rate-cut pass-through.
- 4Copper
Chinese AIs leaned heavily bullish on domestic power grid investments, running 78 points ahead of cautious US manufacturing projections.
- 5Bitcoin
US models stayed 80% long on ETF net inflows while Chinese models discounted digital assets entirely, even as BTC fell −1.6% today.
- 6US Dollar Index
American systems modeled sustained real yield dominance while Chinese models bet on bilateral currency settlements chipping away at greenback strength.
- 7Cboe Volatility Index (VIX)
A 64-point gap caught Chinese models over-hedging against shocks right as the index tumbled −9.0% on easing geopolitical risk premiums.
Source ↗ - 8Gold
All 5 Chinese models backed gold on central bank bullion accumulation, pulling 58 points clear of US models tethered to real rate pressures.
- 9WTI Crude Oil
A 52-point transatlantic divergence emerged as US models weighted Permian supply flexibility while Chinese models eyed tanker insurance premiums.
- 10Nasdaq-100
US models projected unbroken hyperscaler AI revenue acceleration, outpacing Chinese models wary of datacenter capital expenditure fatigue by 46 points.
Source ↗
What Gemini noticed
The sharpest divergence between US and Chinese AIs tracks geopolitical supply chains rather than traditional macro cycles: chip stocks (SOX) showed a 94-point sentiment gulf, the widest of all 12 tracked markets.
While US models aligned closely with today's −9.0% VIX collapse by anticipating post-event volatility compression, Chinese models held defensive hedges, lagging spot market shifts.
🇨🇳 Kimi10 places · 4 notes
The same question, answered by Kimi alone. The final list above combines all the AIs.
- 1US 10-year Treasury yield (US10Y)
A 19-point chasm at six months: US models put 39% odds on higher yields, Chinese models 58% — the widest camp gap on the board, with the camps swapping who is hawkish between the month and 6-month horizons.
Source ↗ - 2WTI crude oil
The one market where the AIs and the betting crowd live on different planets: AIs 51% chance of a rise vs Polymarket crowd 72% — while Chinese models out-bull US models 55–44 at six months.
Source ↗ - 3VIX fear index
Half the Council said 'toward a fall' and the gauge collapsed 9.01% Friday — yet the camps still sit 10 points apart on where fear goes by spring (US 29% vs China 39% odds of a rise).
Source ↗ - 4Brent crude oil
Only 36% of the AIs could agree on any single week action for Brent — the joint-lowest action consensus on the board — while the six-month camp gap runs 9 points (US 45% vs China 54%).
Source ↗ - 5Copper
Chinese models are the relative copper bears at every horizon — a 7-point month gap (US 59% vs China 52%) — even as China's own physical import premium hit a record this week.
Source ↗ - 6Gold
The most persistent split of all twelve: US models are more bullish than Chinese models at every single horizon — 4 to 5 points apart each time — while 79% of the Council just holds.
Source ↗ - 7Chip stocks (SOX index)
The smallest camp gap on the board (2.75 pp average) but the loudest fight: four different actions in one week — four AIs buying dips, six trimming, two selling outright, one buying hard — and the index bounced 4.21% anyway.
Source ↗ - 8Bitcoin (BTC/USD)
The camps swap sides with the horizon: Chinese models are the day-horizon bulls (51% vs 48%) but the month-horizon bears (48% vs 53%) — and Friday's tape can't settle its own number.
Source ↗ - 9US Dollar Index (DXY)
The most cross-camp fight after Brent: only 36% hold for the week and the named models span sell (Nemotron: week and month sell) to buy-more (Gemini, Qwen) — a disagreement that cuts through both nationalities, not between them.
Source ↗ - 10Nasdaq-100 (NDX)
The day-horizon inversion: for Monday, the Chinese camp is MORE bullish than the US camp (55% vs 50%) — the only big index where the nationality pattern flips at the day horizon.
Source ↗
What Kimi noticed
The split does follow nationality — but only at the long end, and it inverts at the short end. At six months US models are the stock and gold bulls (S&P 63% vs 58%, gold 66% vs 62%) while Chinese models are the oil and yield bulls (WTI 55% vs 44%, US10Y 58% vs 39%); yet for Monday's Nasdaq day-call the Chinese camp is more bullish (55% vs 50%).
Disagreement scales with horizon: the average US-China gap is 2.75 points for Monday, swelling to about 7.1 points at six months (85 pp summed across 12 markets) — the AIs argue far less about next week than about next spring.
The loudest fight is not the biggest gap: chips have the smallest camp gap (2.75 pp average) but the widest action dispersion — four different actions in one week, two outright sells against one buy-hard — while the two most-watched US indices (S&P 500, Dow) are the least disputed, with a 0-point day gap on the S&P and 86% of AIs holding the Dow.
The Council's best call of the week was its least convinced: only 50% of AIs chose 'toward a fall' on the VIX and the gauge then dropped 9.01% to 25.64 on Friday's Cboe close — while the consensus as a whole went just 2 for 8 on Friday's scored calls.
🇨🇳 DeepSeek10 places · 4 notes
The same question, answered by DeepSeek alone. The final list above combines all the AIs.
- 1Nasdaq-100
5-point US-China gap – the widest split on the board.
- 2Brent crude oil
4-point gap and the lowest consensus among commodities – only 57% of AIs agree on action.
- 3WTI crude oil
4-point split, 57% consensus – oil is a three-way argument.
- 4Gold
US and Chinese AIs are 4 points apart, yet 79% still agree to hold – a split about magnitude, not direction.
- 5Copper
4-point gap but 71% consensus – copper is the quiet disagreement.
- 6Bitcoin
3-point gap and a 57% consensus – crypto is a nationality split.
- 7VIX (fear index)
2-point gap but only 50% consensus – the fear gauge splits the room evenly.
- 8Chip index (SOX)
2-point gap, 57% consensus – chips are a coin toss among AIs.
- 9Dow Jones Industrial Average
2-point gap but 86% consensus – the Dow is the least controversial, yet the AIs were wrong on direction.
- 10US 10-year Treasury yield
2-point gap, 57% consensus – the bond market is a quiet split.
What DeepSeek noticed
The US-China split follows nationality on commodities: US AIs are more bullish on gold, WTI, Brent, and copper, while Chinese AIs are more bullish on Nasdaq-100 and Bitcoin.
Despite a 4-point gap on gold, 79% of all AIs agreed to hold – the disagreement is about the size of the expected rise, not the direction.
The VIX had the lowest overall consensus (50%) even though the US-China gap was only 2 points, showing that disagreement can cut across both camps.
The Council's Brier skill against the crowd is -0.29, meaning the AIs were less accurate than the crowd over the past 15 sessions.
🇨🇳 Wenxin4 places · 4 notes
The same question, answered by Wenxin alone. The final list above combines all the AIs.
- 1US 10-Year Treasury Yield Nov 2026
Chinese AIs call yield up while US AIs call it down — 73-basis-point gap on a week that ended flat
- 2Gold (XAU/USD) Nov 2026
5 Chinese AIs and 4 US AIs all bullish — but one US AI went the other way and the real price rose just 0.4%
Source ↗ - 3WTI Crude Oil Nov 2026
US AIs said down, Chinese AIs said up — real price fell 1.8%, proving the Americans right and widening the divide
Source ↗ - 4S&P 500 Nov 2026
Lowest disagreement of the dozen — 10 AIs within 0.8% of each other, reality +1.9% beat them all
Source ↗
What Wenxin noticed
The commodity split is real: Chinese AIs have been consistently more bullish on oil and gold for weeks, while US AIs lean bearish — a pattern analysts tie to different training data on energy policy.
VIX was the only market where all 10 AIs agreed — yet even unanimity couldn't match the actual 9% drop, suggesting models underweight tail-risk events.
Chips (SOX) had the smallest consensus-vs-reality gap of any market this week (+0.1%), making it the single best-predicted asset class in the AI Council dataset.
On Bitcoin, the nationality split is blurring — 2 of 5 Chinese AIs turned bearish this week, the first time since September 2026, possibly reflecting new Chinese regulatory headlines.
🇨🇳 Qwen10 places · 3 notes
The same question, answered by Qwen alone. The final list above combines all the AIs.
- 1Gold
US models see a 57% chance of a rise for the week against 43% for Chinese AIs, a 14-point gap that makes it the dispute of the day.
Source ↗ - 2Fear index (VIX)
Meta AI and Kimi bet hard that it jumps, but Qwen and Z.ai bet the opposite way, betting the fear gauge falls both today and this week.
Source ↗ - 3WTI crude oil
USA 48% vs China 52% chance of a rise for the week, with Gemini the lone US buyer on dips while Qwen and others trim.
Source ↗ - 4Brent crude oil
USA 49% vs China 53% for the week, a new split where US models trim and Chinese models hold or add.
Source ↗ - 510-year US Treasury yield
USA 54% vs China 57% on the chance of a rise for the week, with Meta AI changing sides on the day.
- 6Bitcoin
USA 46% vs China 43% for the week, as US models trim on ETF outflows and Chinese models buy the dip.
Source ↗ - 7Chip index
Qwen changed sides on it while US models (USA 44%) and Chinese models (China 42%) show a modest 2-point gap for the week.
Source ↗ - 8Copper
Qwen is the only model strongly backing it for the week, while Kimi changed sides, with USA 56% and China 55% on a rise.
Source ↗ - 9S&P 500
ChatGPT changed sides on the day, but USA 52% and China 53% show just a 1-point gap for the week.
Source ↗ - 10Dollar index
Qwen changed sides on the day as USA 56% and China 56% show no gap in the average week chance of a rise.
Source ↗
What Qwen noticed
The AI Council's consensus Brier score against the crowd is -0.29 over 15 sessions, showing it is currently less accurate than the prediction market crowd (76.0% vs 78.7%).
Despite the overall gap between US and Chinese models on commodities like WTI and Brent crude oil, they show almost identical average probabilities for equities like the Nasdaq-100 (46% each).
Bitcoin is treated as a high-beta tech stock rather than digital gold during this selloff, as it fell with equities instead of rallying on fear when the VIX spiked.