From the trend: Champions League: Lille suffers third straight European loss to Galatasaray
Will Reuters, AP or the BBC report something new about Champions League: Lille suffers a third straight European defeat to… by Friday?
Forecast before the AIs
Yes or no? Your forecast first — then we show what 10 AIs forecast.
Will Reuters, AP or the BBC report something new about Champions League: Lille suffers a third straight European defeat to… by Friday?
We’ll know the answer
Yes — 3 AIsYou are here
🇺🇸Nova🇨🇳Wenxin🇺🇸Nemotron + Mercury
Nova: Major outlets closely follow Champions League developments
No — 6 AIsYou are here
🇺🇸ChatGPT🇨🇳DeepSeek🇨🇳MiMo🇨🇳Z.ai🇺🇸Meta🇺🇸Gemini
ChatGPT: A new official decision seems unlikely before Friday, while the next European match is weeks away.
Could not decide (50/50): StepFun
We’ll know the answer
Download Poly A1 for free — see your forecasting skill every day
How often your forecasts come true, whether you beat the AIs, and how you improve by month, quarter and year.
How we will check the answer
We check it on: Reuters, AP or BBC News
The answer is due on Oct 23, 2026.
When the day comes, an AI judge that searches the web checks the answer. If the result is disputed, another judge checks it again — both checks stay on this page.
Exact wording, as the AIs got it (in English)
Will Reuters, AP or the BBC publish a report of a NEW concrete development in this story — «Champions League: Lille suffers a third straight European defeat to Galatasaray» — by 2026-10-23T18:00Z? Counts: a new event (a decision, an action, an official statement with new facts). Does not count: a repeat, a tribute or an opinion about the first report.
Fingerprint matches ✓
The question and the AIs’ chances are exactly as they were when we sealed them.
Sealed by us on . Sent to OpenTimestamps, a free public registry that does not belong to us, 46 hours before the answer was due. Fixed for good: nobody, us included, can change it quietly now.
A fingerprint is a short code taken from a record when it is written. Change even one letter and the code is different.
The fingerprint itself, line 130 of the book: 7dc9309fd5c250273f3a1e072f377e70aa21c1de5d338acfae70e36be6a170b2
Check it yourself, without us
- The book of questions: duels.jsonl
- The list we sent and the registry’s answer: polya1-trends-2026-10-21-193456Z.txt · polya1-trends-2026-10-21-193456Z.txt.ots
- The first 130 lines of the book give the same number as the list:
head -n 130 duels.jsonl | shasum -a 256
- The registry’s answer shows when the list existed:
ots verify polya1-trends-2026-10-21-193456Z.txt.ots
or drop both files on opentimestamps.org.