AI Evaluation Intern - Law & Regulatory

We're hiring law and regulatory master's students to evaluate AI-generated legal and compliance content. This is a 12-week, full-immersion internship. The work is technical, hands-on, and paid. Around 20% of your time goes toward structured AI training: how large language models work, how RLHF fits into the development pipeline, what good evaluation looks like and why it matters. No prior AI knowledge is required. You'll learn everything on the job. The other 80% is live evaluation work on production AI models. This is the core of the internship. Your legal knowledge and your understanding of EU regulatory frameworks are the reason you're here.

KEY RESPONSIBILITIES 1. Evaluating AI-generated legal and regulatory content. You'll review AI outputs on EU regulations (GDPR, EU AI Act, Digital Markets Act, Digital Services Act, CSRD) and score them for accuracy. These aren't legal opinions. They're factual questions: what does the regulation say, how does it apply, and what are the key compliance concepts. You'll check whether the AI's reasoning chains are logically sound and whether the regulatory references are real. AI labs building compliance tools and enterprise knowledge assistants need this evaluated by people who've studied the material, and US-based competitors don't have evaluators with embedded EU regulatory knowledge. 2. Fact-checking AI in your native language. AI models hallucinate. They cite articles that don't exist, misstate regulatory thresholds, and confidently get jurisdictional details wrong. You'll catch them. If you speak French and English, you might review legal AI outputs in both languages, flagging incorrect citations, misapplied rules, and translations where the legal meaning shifts. Most AI evaluation today only covers English. European legal content in other languages is almost entirely unevaluated. 3. Red-teaming AI for safety and compliance. The EU AI Act now requires companies deploying high-risk AI systems to test their models for harmful outputs. You'll try to break them. That means finding prompts that lead to incorrect legal advice, biased outputs, or dangerous recommendations, documenting exactly how the model fails, and scoring the severity. This is becoming a compliance requirement, not optional, and regulators are watching. 4. Building regulatory QA datasets. AI labs training legal and compliance assistants need structured question-answer pairs written by people who understand the source material. You might write and evaluate QA pairs covering GDPR data processing requirements, EU AI Act provider obligations, or Digital Markets Act compliance concepts. Each answer includes a reasoning chain explaining the regulatory logic. This isn't practicing law. It's explaining regulations clearly, which is core curriculum in European business law programs. 5. Contributing to industry benchmarks. The evaluation work you do here feeds directly into the benchmarks top-tier AI labs use to assess and publicly report the factual capabilities of their models. When OpenAI, Google DeepMind, or Anthropic measure how well their models reason about legal and regulatory subjects, datasets built by evaluators like you are part of what they test against. IDEAL QUALIFICATIONS Currently enrolled in a Master's in Corporate Law, Business Law, EU Law, Regulatory Affairs, or a closely related program Fluent in English plus at least one other European language (German, French, and Spanish are in highest demand) Comfortable working independently in a fully remote setup Reliable internet connection and a quiet workspace NICE TO HAVE Prior experience in corporate law, compliance, regulatory affairs, or legal consulting Practical exposure to GDPR, EU AI Act, or other EU regulatory frameworks Experience working across cultures or in international teams CONTRACT & PAYMENT TERMS

Depending on experience (paid)