KellyBench recently conducted a study examining the effectiveness of various AI models in the realm of sports betting during a full Premier League season. The models tested included Claude, GPT-5, Gemini, and Grok. Despite high expectations, none of these AI systems managed to yield a profit throughout the season.
The significance of this study lies in its exploration of the intersection between artificial intelligence and sports wagering. As the betting industry continues to seek new strategies, AI tools promise insights that could influence betting decisions. This research highlights the current limitations of AI in accurately predicting sports outcomes, raising questions about the reliability of these technologies in a competitive market.
Throughout the testing period, the four models engaged in various betting scenarios. Analysts had anticipated that at least one would demonstrate profitability, given the advancements in AI technologies. However, the results showed a uniform inability to generate returns. This result casts doubt on the current capabilities of AI in sports betting, especially in a league as unpredictable as the Premier League, known for its competitive matches and upsets.
As the betting market evolves, attention will focus on future developments in AI technology and its applications in sports analytics. Upcoming betting events, including the Champions League fixtures, may provide new opportunities for these models. Tracking their performance in these high-stakes environments will be crucial for understanding their viability in predicting outcomes. A specific price level for betting markets to observe will be when the next set of odds are released for these fixtures.
