What Enterprise AI Gets Wrong About Legal Operations

Enterprise AI vendors have discovered the legal market. The pitch is consistent: automate document review, accelerate research, reduce overhead, scale without adding headcount. For a large firm with a dedicated technology team and a six-figure software budget, some of that pitch is relevant.

For a solo immigration practitioner or a three-attorney practice managing 300 active matters, almost none of it is.

The Scale Mismatch

Enterprise legal AI is designed for the problems of large organizations: managing massive document sets across complex litigation, coordinating workflows across dozens of timekeepers, maintaining consistency across practice groups in different cities. Those are real problems. They are not the problems of a small immigration practice.

The problems of a small immigration practice are different. Manual intake that requires re-entry across multiple systems. Follow-up sequences that live in the paralegal's head rather than a documented process. Deadline tracking in a spreadsheet that nobody fully trusts. Client communication that is inconsistent because there is no system enforcing consistency. Enterprise tools do not address those problems. They add complexity on top of them.

The Compliance Assumption Gap

Enterprise AI is also built with assumptions that do not map to immigration practice. The compliance constraints in immigration work are specific, practice-area-level constraints that a general-purpose legal AI tool will not have been designed around.

What can be automated and what cannot in an immigration context is not the same as what can be automated in a contract review or due diligence context. Immigration work involves credibility assessments, humanitarian considerations, and practice-specific regulatory limits on what non-attorney staff can do. A tool that was not designed with those constraints in mind will not enforce them.

What Small Immigration Practices Actually Need

What works for a small immigration practice is narrow, deliberate, and compliance-aware. A well-designed intake system that handles the specific matter types the practice manages. A follow-up workflow that runs automatically without requiring manual supervision. A deadline visibility system that surfaces what is coming without requiring someone to maintain a separate spreadsheet.

These are operational problems that require operational solutions, built by someone who understands how immigration practice actually works. That is a different product from enterprise AI. It is also a more useful one.