A recent analysis tested the AI model Jev against 26,174 resolved Polymarket markets to determine if it could predict market outcomes better than the market price itself. The study found that Jev performed poorly, achieving an AUC of 0.615 on markets resolved after its release, significantly underperforming the 0.844 AUC of the market price from a day prior. Furthermore, providing Jev with the market price information actually worsened its performance, indicating a lack of predictive capability beyond random chance. AI
IMPACT This analysis suggests that current AI models may not yet surpass established market prediction mechanisms, indicating limited immediate impact on automated trading or prediction platforms.
RANK_REASON The item describes a test of an AI model's performance on a specific task, which falls under the 'tool' category as it evaluates a product's capabilities.
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →