// AI news · 2026-10-29 · WIRED
AI Agents Manage Fantasy Soccer Team, Still Outperformed by Humans
AI agents managing a fantasy soccer team still underperform humans.
Published 2026-10-29 02:30 (Beijing time)Source WIREDAuto-summarised, fact-checked against source
01Summary
Early suggestions made flawed decisions like benching Haaland. A structured multi-agent system—reading news, analyzing results, modeling, and reviewing—improved outcomes significantly. GPT-5 outperformed others in prediction tasks. While AI matches or beats top human forecasters in some cases, it still lacks deep understanding of real-world dynamics behind events.
Summarised from the original article by an open-source model, then checked automatically: every number and English name must appear in the source. The original article is authoritative.
02Key points
- Initial team suggestions incorrectly benching Haaland show poor real-world reasoning
- Structured multi-agent system (news, results, modeling, review) improves team quality
- GPT-5 performs best in prediction tasks, followed by Grok 4 and Claude Sonnet
- AI still lacks deep understanding of underlying dynamics in real-world events
03Source
I Put a Bunch of AI Agents in Charge of a Fantasy Soccer Team. They’re Not Better Than Humans—Yet
WIRED · published 2026-10-29 02:30 Beijing time
05Related news
- Microsoft Reports $10B Revenue from OpenAI in Q32026-10-29 · TechCrunch AI
- Oracle Announces Cloud and AI Investment Plan2026-10-29 · IT之家
- OpenAI Claims Models Refute ETH2026-10-28 · TechCrunch AI
- OpenAI Launches Always-On Personal Assistant Dots2026-10-27 · TechCrunch AI
- 41,000 DHS Workers' Data Leaked2026-10-29 · Wired
- GitHub Copilot Fusion supports local model routing2026-10-28 · github-changelog-copilot
From AIMap AI news: collected from 1,100+ official and media sources. Summaries only restate the original article; follow the link for the full story. All AI news · Today