Contact Legal AI Research Experts for Inquiries

Welcome Your Questions

We welcome careful questions, practical feedback, and direct notes from people working around law, data, public decision-making, and automation.

Some messages arrive with a narrow technical point: a question about model evaluation, a missing citation, or a concern about how legal prediction work is described. Others are broader. A policy researcher may be trying to understand where automated reasoning helps and where it becomes risky. A student may be looking for a plain explanation of a challenge result. A practitioner may want to know whether a proposed use of AI in legal work has been tested with enough care.

Those are all suitable reasons to write.

For general contact, use [email protected]. A short message with context usually helps more than a long attachment. If you are asking about a specific article or research theme, include the page title or topic in the subject line.

General Business Inquiries

Business inquiries tend to move faster when the first email explains the practical question, not just the category of interest.

We read for substance. A clear two-paragraph note is often enough: the first paragraph explains the matter, and the second says what response would be useful. If the request touches privacy, user data, or site terms, please review the Privacy and Terms pages before writing. That saves everyone a round of clarification.

We do not publish phone numbers or physical mailing addresses for unsolicited inquiries.

Press and Media Contacts

Journalists usually need three things: a timely reply, a quotable explanation, and a boundary around what can be said responsibly. We try to keep all three in view.

Media questions may concern legal AI performance claims, prediction challenges, access to justice, consumer claims processes, or the limits of automation in public and private legal systems. Please name the outlet, format, publication date, and whether you need background comment, attributed comment, or a short technical explanation.

Short-notice requests are easier to answer when the scope is narrow. For example, “Can legal AI predict outcomes?” is too broad for a useful same-day reply. “What should readers look for when a vendor claims high accuracy on legal outcome prediction?” is the kind of question that can be handled with more precision.

We may decline requests where the format would require oversimplifying a technical or legal point. That is not reluctance to comment; it is a preference for being accurate over being decorative.

Partnership Opportunities

Partnership conversations work best when they begin with a shared research or public-interest question.

Case Crunch is interested in serious work on legal AI evaluation, automated reasoning, consumer claims, justice automation, and methodology. That may include academic discussion, challenge design, educational material, data governance review, or carefully scoped public communication. It does not need to be large to be worthwhile. Some of the most useful exchanges begin with a single dataset problem or a disagreement about how a benchmark should be interpreted.

What makes a proposal easier to assess

Describe the problem, the people involved, the expected output, and the decision points. If data is involved, say what kind, who controls it, whether it includes personal information, and what ethical or legal constraints already apply. If the proposal concerns public-facing claims about legal AI, include the evidence you expect those claims to rest on.

Send partnership notes to [email protected]. If the idea is early, say so. A rough but honest outline is better than a polished proposal that hides the hard parts.

For background on the project and its areas of interest, visit the About page.