Stanford study finds AI outperforms law professors on legal tasks

Researchers tested GPT-4 and other AI models against law professors on contract review and legal reasoning tasks, with the models consistently scoring higher. The finding underscores a real shift in how legal work might be structured—but doesn't mean replacing lawyers; rather, it highlights where AI tooling could handle routine analysis faster, freeing professionals for judgment calls that still need human judgment.