Legal AI training jobs pay attorneys to grade and improve the legal reasoning AI models produce: contract analysis, case synthesis, statutory interpretation, jurisdiction-specific advice. Labs cannot ship a model that hallucinates case law, so they pay lawyers to catch it. The work is remote, hourly, and screened on real legal training rather than AI skills.
Every live law listing from Mercor, Outlier, Alignerr, micro1, DataAnnotation and Terac is below, refreshed daily, with rates as published.
105 open law listings live right now.
Filtering by rate hides per-task listings, which have no hourly equivalent.
Write original multiple-choice questions on Swedish law for an AI evaluation dataset. You will draft exam-style questions from your own professional expertise, each with ten answer…
Write original multiple-choice questions on Hungarian law for an AI evaluation dataset. You will draft exam-style questions from your own professional expertise, each with ten answ…
Write original multiple-choice questions on Thai law for an AI evaluation dataset. You will draft exam-style questions from your own professional expertise, each with ten answer op…
Write original multiple-choice questions on Danish law for an AI evaluation dataset. You will draft exam-style questions from your own professional expertise, each with ten answer…
Write original multiple-choice questions on Finnish law for an AI evaluation dataset. You will draft exam-style questions from your own professional expertise, each with ten answer…
Write original multiple-choice questions on Taiwanese or Hong Kong law, in Chinese (Traditional), for an AI evaluation dataset. You will draft exam-style questions from your own pr…
Write original multiple-choice questions on the law of your jurisdiction, in Arabic, for an AI evaluation dataset. You will draft exam-style questions from your own professional ex…
Write original multiple-choice questions on Chinese law for an AI evaluation dataset. You will draft exam-style questions from your own professional expertise, each with ten answer…
Join a growing network of industry experts supporting AI research and development. Overview We're building a select group of experienced legal professionals to help evaluate and im…
Mercor is seeking experienced Pro Bono Counsel to evaluate AI-generated legal analyses involving civil legal services and access-to-justice issues. Experts will leverage their broa…
Mercor is seeking experienced Senior Civil Legal Paralegals and Accredited Representatives to review AI-generated legal guidance involving civil legal services workflows. Experts w…
Mercor is seeking experienced Clinical Law Professors and Clinic Directors to evaluate AI-generated legal reasoning involving civil legal services. Experts will combine doctrinal k…
Mercor is seeking experienced Public Interest Attorneys to evaluate AI-generated legal reasoning across a broad range of civil justice matters. You'll review realistic legal scenar…
Projects run from redlining a model’s contract review, to writing model answers for bar-style essays, to checking whether cited cases exist and say what the model claims. Some work is drafting: producing the well-reasoned memo a model should have written. Some is adversarial: constructing fact patterns where the tempting answer is wrong.
US-qualified attorneys see the most volume, and listings often specify a practice area, corporate, litigation, IP, tax, or a jurisdiction. Practising lawyers moonlight here; so do former practitioners, judicial clerks and LLMs-turned-JDs. What gets screened is whether your reasoning survives being read by another lawyer.
Six marketplaces publish this work openly. Every listing above links to the one that posted it, and each has its own page here covering pay, screening and who gets hired.
Every marketplace above screens with an interview or assessment in your own domain before assigning paid work. Run that interview with Skillora first and get feedback on how you explained your reasoning, not just on what you said.
Practise the screen free
