AI Research Legal Counsel

EngineersOfAI; San Francisco (CA); Hybrid; USD 180,000 - 260,000; Full time

Job summary

Anthropic is seeking a Research Legal team member to advise on legal topics across the lifecycle of frontier AI models, including data acquisition, training, evaluation, and deployment. You will be a subject matter expert on copyright, privacy, data use, and AI regulation and work with researchers directly on unresolved questions.

The ideal candidate will navigate complex legal questions in a fast-moving technical environment, enabling responsible research and product development while

Qualifications

• JD with active state bar admission (California or New York preferred).
• Ability to apply legal requirements to fast-changing, technical work to enable responsible research.
• Excellent communication and analytical skills with the ability to translate complex issues into actionable guidance.
• Familiarity with EU AI Act, U.S. legislation and AI frameworks (GDPR/CCPA) and global regimes.

Responsibilities

• Advise on data sourcing, licensing, partnerships, and use, shaping data acquisition and use strategy as laws evolve.
• Coordinate regulatory readiness and model launches, review documentation and system cards, respond to inquiries, advise on pre-release and internal models.
• Support Responsible Scaling Policy on biosecurity risk, including evaluations and dual-use questions.
• Counsel multiple research teams on design, data collection/sharing, external collaborations, and publication review.
• Monitor global AI, privacy, copyright, and export-control developments and translate into internal guidance with stakeholders.
• Engage with regulators and industry groups to develop norms and best practices for responsible AI and data governance.

Skills

Legal analysis

Policy analysis

Communication skills

Technical fluency

Regulatory knowledge

Stakeholder collaboration

Education

Juris Doctor (JD)

Job description

About Anthropic

Anthropic’s mission is to create reliable, interpretable, and steerable AI systems. We want AI to be safe and beneficial for our users and for society as a whole. Our team is a quickly growing group of committed researchers, engineers, policy experts, and business leaders working together to build beneficial AI systems.

About the role

As a member of the Research Legal team, you'll advise on legal topics across the full lifecycle of building and studying frontier AI models: data acquisition and use, training pipelines, model evaluation and launch, and alignment, interpretability, model welfare, societal impacts, and beneficial deployments research, as well as Responsible Scaling Policy work on biological risk. You'll be a subject matter expert on copyright, privacy, data use, and AI regulation (including the EU AI Act, U.S. state laws, and emerging global rules), and will work directly with researchers on questions that often have no settled answer.

The ideal candidate will have navigated complex legal questions at the intersection of machine learning, data, and academic research in a fast-moving technical environment, with the judgment to work at the cutting edge of AI development while effectively solving issues, enabling innovation, and calibrating risk. This is a uniquely positioned role touching many different facets of AI, presenting the opportunity to be a leader in AI legal thought.

Key responsibilities

• Advise research and data teams on data sourcing, licensing, partnerships, and use, and help shape Anthropic's data acquisition and use strategy as laws and markets evolve
• Help coordinate regulatory readiness and model launches, review model documentation and system cards, respond to regulatory inquiries, and advise on legal questions for pre-release and internal models
• Support the Responsible Scaling Policy as applied to biological and chemical risk, including capability evaluations, tiered access for biosecurity researchers, and dual-use research questions
• Counsel the Societal Impacts, Economic, Alignment, Interpretability, Model Welfare, and Beneficial Deployments teams on research design, data collection and sharing, external collaborations, program launches, and publication review
• Monitor global AI, privacy, copyright, and export-control developments and translate them into practical internal guidance and policy, working closely with Product, Policy, and other legal stakeholders
• Engage with academics, regulators, industry groups, and civil society to help develop norms and best practices for responsible AI development and data governance

Minimum qualifications

• Hold a law degree (JD, LLB, or equivalent) and meet the local licensing requirements to work as in-house counsel where you'll be based (California or New York preferred)
• Practical, creative problem-solving: the ability to apply legal requirements to fast-changing, deeply technical work, with an understanding that the goal is to enable responsible research and product development in a risk-calibrated environment
• Exceptional communication and analytical skills, technical fluency, and the ability to navigate ambiguous and uncharted legal territory, turning complex technical and legal analysis into advice people can act on
• Familiarity and comfort advising on the EU AI Act, U.S. legislation and AI frameworks, GDPR/CCPA, and comparable regimes globally, sometimes where no AI-specific legislation exists
• Interest in thinking deeply about the impact of AI on society and how to make humanity's transition to the AI age as beneficial as possible
• Sound judgment about when to be flexible and when to hold a line; comfort making high-stakes calls in unsettled areas and outside your core expertise; and a low-ego, collaborative approach, doing the work needed rather than delegating it or being territorial

Preferred qualifications

• At least 8 years of legal experience advising technology companies on copyright, privacy, data, AI regulation, and related product and policy issues, ideally including time in-house and work on AI/ML products or services
• Experience engaging directly with regulators or standards bodies
• Working technical literacy: enough familiarity with how models are trained, evaluated, and deploy