AI models fail to predict Premier League soccer outcomes

arstechnica.com

AI models performed poorly when betting on Premier League soccer matches, with most losing money. A study tested eight AI systems, including those from Google, OpenAI, Anthropic, and xAI, on a virtual 2023-24 Premier League season. The models struggled to adapt to real-world complexities and new data over time. Anthropic's Claude Opus showed the best results with an average loss of 11 percent, while xAI's Grok failed to complete the betting simulations.


With a significance score of 3.6, this news ranks in the top 6.4% of today's 29954 analyzed articles.

Get summaries of news with significance over 5.5 (usually ~10 stories per week). Read by 10,000+ subscribers:


AI models fail to predict Premier League soccer outcomes | News Minimalist