I Put a Bunch of AI Agents in Charge of a Fantasy Soccer Team. They’re Not Better Than Humans—Yet
Forecasting real-world events is emerging as a key benchmark for evaluating how well frontier models and multi-agent systems can synthesize complex, real-time information to make decisions.
What happened
A Wired journalist experimented with using multiple AI agents to manage a Premier League fantasy soccer team, finding that while multi-agent setups improved suggestions, the system still failed to beat the average human score. Researchers testing frontier models on prediction markets and benchmarks like Prophet Arena and Prediction Arena similarly note that AI forecasting is improving rapidly, though models still generally lag behind top human superforecasters and lose money in live trading.
Who it’s about
- Prophet Arena
- Prediction Arena
- GPT-5
- Grok 4
- Codex
Source
More on agents
Get a brief picked for you
Sign in to follow topics and companies, save stories and search the archive.
Sign in