AI做科研？最强模型只拿了21分
我们总说AI能帮科学家干活，但让它从头到尾独立完成一项研究呢？这篇论文给了一个残酷的答案：目前最强的AI科研助手，满分100分只拿了21.5分。研究者设计了一套测试，涵盖10个科学领域的40个真实课题——每个课题都基于一篇已发表论文，给AI提供原始数据和相关文献，但隐藏目标论文，看它能否独立“重新发现”论文中的结论。结果发现，AI最常栽在实验方案不对、证据对不上、以及漏掉核心科学问题。它不是你明天能用上的，但告诉你：AI离真正的自主科研还差得远。
📄 原文摘要(英文)
AI coding agents are increasingly used for scientific work, but their end-to-end autonomous research capability remains difficult to verify. We present ResearchClawBench, a benchmark for evaluating autonomous scientific research across 40 tasks from 10 scientific domains. Each task is grounded in a real published paper, provides related literature and raw data, and hides the target paper during evaluation. Expert-curated multimodal rubrics decompose the target scientific artifacts into weighted criteria, enabling evaluation of target-paper-level re-discovery while leaving room for new discovery. We evaluate seven autonomous research (auto-research) agents under a unified protocol and seventeen native LLMs through the lightweight ResearchHarness. Current systems remain far from reliable re-discovery: the strongest autonomous agent, Claude Code, averages 21.5, and the strongest ResearchHarness LLM, Claude-Opus-4.7, averages 20.7, with an LLM frontier mean of only 26.5. Error analysis shows that failures concentrate in experimental protocol mismatch, evidence mismatch, and missing scientific core. ResearchClawBench provides a reproducible evaluation frontier for measuring progress toward autonomous scientific research.