Advanced AI models are proving remarkably poor at betting on soccer matches, with a recent study concluding they systematically underperform human predictors. Every frontier model evaluated lost money over the season, with many experiencing complete financial ruin.
This finding offers a counterpoint to widespread anxieties about AI's potential to displace human jobs across various sectors. Researchers noted that many common AI benchmarks are flawed, tested in static environments unlike the chaotic real world.
Ross Taylor, a study author and General Reasoning CEO, stated, "There is so much hype about AI automation, but there’s not a lot of measurement of putting AI into a longtime horizon setting." He emphasized that while AI excels in some tasks like software engineering, it struggles with real-world activities requiring longer-term strategic thinking.