DigestAI news desk

Cut through the AI noise.

Agents & Tools9 min read

Microsoft Copilot AI predicts NFL Week 3 results

Microsoft Copilot AI has kept a perfect 11‑5 record in each of the first two weeks of the 2026 NFL season, giving it a 22‑10 overall record so far. USA TODAY Sports continued the experiment by prompting the chatbot to predict all 16 Week 3 games, using the query “Can you predict the results of every Week 3 game for the 2026 NFL season and explain your logic for each prediction?”.

1 source

Key points

  • Copilot posted an 11‑5 record in each of the first two weeks, totaling a 22‑10 season record.
  • USA TODAY fed Copilot the prompt to predict all 16 Week 3 games, receiving a full set of picks.
  • The AI correctly chose all three prime‑time games in Week 2 but sometimes gave outdated info that required correction.

The AI generated picks for every matchup, correctly selecting all three prime‑time games in Week 2 and backing several home teams. Occasionally it supplied outdated information, which the editors corrected by asking Copilot to fix the errors. Human analysts added brief commentary to each prediction, noting injuries, player performances and potential mismatches.

Overall, the exercise shows that large language models can be applied to sports forecasting, but their need for real‑time data updates remains a limitation.

Full story from usatoday.com · by Jacob Camenker · via Search: MicrosoftOpen source ↗

NFL Week 3 predictions by Microsoft Copilot AI for every game

usatoday.com · 24 September 2026

Loading the full article…

This text was published by usatoday.com and written by Jacob Camenker. It is reproduced here with attribution so you can read it in full; the rights remain with the publisher. Read it at the source ↗

Topics · follow one to build your own front page
MicrosoftUSA TODAYMicrosoft Copilot AIZach Bako‑BeweleMichael Penix Jr.Josh AllenBryce YoungDeshaun WatsonC.J. Stroud

The headline, key points and digest above were generated by Digest AI's editorial model from the linked sources. Automated summaries can contain errors: the sources are the record. Spotted a mistake? Tell us. Published by Martin K., who runs Digest AI and handles corrections.

Comments

via GitHub Discussions

More in Agents & Tools

All →

Related stories