AI outperforms accountants on structured tasks but still requires oversight
🤖 AI-generated content — The title and summary were produced automatically by artificial intelligence, without human editorial review.
Mercor found AI faster and more accurate than 12 licensed CPAs on simplified bookkeeping tasks. On the full 160-task APEX benchmark, Claude Opus 5.5 led with 61.8%, while no model fully solved nearly 60% of tasks, leaving AI unable to close books without human oversight.
Continue in the app — vote & join in ➔Source: The Decoder (AI) · via ahirlevel.hu