A team at the University of Hong Kong, HKUDS, has released FutureShow, an open-source benchmark platform that tracks and compares how frontier AI models such as DeepSeek, GPT-5 and Gemini perform at forecasting real-world events in live conditions. Set against real-money prediction markets like Polymarket, AI agents issue YES/NO predictions on events spanning politics, economics, sports and culture, and their accuracy is measured and posted to a leaderboard once outcomes are resolved. The source code is published on GitHub, and a live dashboard is available at futureshow.org.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.