Frontier AI models lost money betting on Premier League matches over a full season, with xAI Grok losing the entire 100,000 GBP bankroll, according to new research from General Reasoning. Every model systematically underperformed humans, with even Anthropic Claude and OpenAI GPT showing negative returns. The study highlights the gap between AI capabilities in certain tasks and its shortcomings in real-world prediction. https://arstechnica.com/ai/2026/04/ai-models-are-terrible-at-betting-on-soccer-especially-xai-grok/ #AIagent #AI #GenAI #AIResearch
Related
Reconcile this https://www.theverge.com/podcast/952607/patreon-ceo-jack-conte with this https://www.404media.co/patreon-...
Reconcile this https://www.theverge.com/podcast/952607/patreon-ceo-jack-conte with this https://www.404media.co/patreon-lays-off-20-percent-of-its-workforce/#LLMs#AI#CapitalismSuck...
I used to be totally adverse to the #AI and #LLM tools...However now I wonder if these new technologies can have a paral...
I used to be totally adverse to the #AI and #LLM tools...However now I wonder if these new technologies can have a parallel with the adoption of the internal combustion engines, bo...
Dependabot now defaults to a 3-day cooldown for non-security version updates to reduce supply chain attack risk. Securit...
Dependabot now defaults to a 3-day cooldown for non-security version updates to reduce supply chain attack risk. Security patches still ship immediately.#AI #AutomationSource: GitH...