90% of US working hours are spent on tasks not covered well by AI benchmarks
mercor.comspeckx
A study finds that 90% of US working hours involve tasks that current AI benchmarks do not adequately measure or assess. This reveals a significant gap between what AI evaluation systems test and the actual work performed across American industries.
Why it mattersAI tool developers and procurement teams use benchmarks to evaluate AI capabilities, so this gap means they may overestimate how ready AI is for real workplace tasks and make poor decisions about AI adoption.
Read at mercor.com