Google accused of using novices to fact-check Gemini's AI answers

Google has altered Gemini's evaluation process, instructing contract workers to rate all prompts, including those outside their expertise. This change raises concerns about evaluation accuracy. Workers are told to rate what they understand and acknowledge knowledge gaps. Google maintains this won't affect Gemini's accuracy, as raters can still provide feedback on format and style. The company recently introduced a new benchmark for checking AI response accuracy.