Mercor study: AI beats accountants on bookkeeping speed and accuracy
What happened
On 2 October 2026, The Decoder reported that Mercor published a study in which 12 certified public accountants (CPAs) with an average of five and a half years of experience completed simplified tasks from the APEX Accounting Benchmark and were compared with AI models. The results showed that AI models were faster, more accurate and cheaper than these accountants on structured bookkeeping tasks. The report also noted that AI still cannot complete closing independently. The study is the only report on this matter so far, and no figures or claims contradicting earlier reports have appeared.
Written by AI from the reports · updated 14 d ago
Timeline
Follow the reports to see the event from each side.
- The DecoderAI beats licensed accountants on speed and accuracy, but still can't close the books without supervision
According to a Mercor study, current AI models now outperform licensed CPAs on structured accounting tasks in both speed and accuracy. Eighteen months ago, they still lagged far behind. On the more demanding APEX Benchmark, though, no model fully completes all the tasks. Without human oversight, AI still can't close the books on its own.