AI Chatbots' Inaccuracies in Financial Queries Highlighted in FT Report
A report by the Financial Times has revealed that popular AI chatbots, including ChatGPT, Claude, Copilot, Grok, and Gemini, provided incorrect answers to financial queries an average of 57% of the time. This figure rose to 88% for more complex queries requiring multiple calculations, with some models being wrong up to 99% of the time. The report also highlighted instances where AI models ignored upcoming changes in tax law and invented new rules, potentially leading to significant financial losses.5 sourcesSee all sources
