Surge AI has released DAYJOB: Tax, a new benchmark designed to evaluate AI agents on professional tax assignments. The benchmark includes 50 environments across eight areas of tax practice, testing agents on their ability to handle complex client requests, supporting records, and make professional judgments. In initial tests, Claude Sonnet 5.5 achieved the highest pass rate at 10.8%, followed by Claude Fable 5.1 and Claude Opus 5.5, with performance varying in relation to cost. AI
IMPACT This benchmark will help assess AI's capability in specialized professional domains like tax advisory, potentially guiding future development for AI agents in complex decision-making roles.
RANK_REASON New benchmark released for evaluating AI agents on professional tasks.
Read on Hacker News — AI stories ≥50 points →
AI-generated summary · Google Gemini · from 2 sources. How we write summaries →