← All journal entries

J/02 · April 2026 · 5 min read · Thesis

Against the innovation committee.

A short brief on why every AI-transformation program at a large firm looks the same - and produces the same nothing.

The innovation committee is a specific kind of institutional gesture. Its purpose is to be seen doing something about a threat without actually doing anything that would threaten anyone in the room.

In BigLaw, that gesture takes an almost identical shape across firms: a Chief Innovation Officer with no P&L, a legal-tech budget line under general expenses, a quarterly pilot with a vendor, and a press release. Nothing that appears in that list can, structurally, change the practice. That is the point.

Change would require the firm to rewrite its billing model, its promotion pipeline, or its compensation formula. Any one of those requires a partnership vote against short-term partner income. Those votes do not happen. Innovation committees exist so that not happening looks like happening.

None of this is a moral failing. It is a governance geometry. A partnership of three hundred owners will always converge on decisions that protect the median owner's income for the next twenty-four months. AI-native rebuilding is a five-year decision that costs the median owner in the first eighteen. It will not survive contact with the vote.

So the AI-native firm gets built somewhere else, by someone else - often a lawyer who left to start their own thing, or a firm that never grew big enough to need a vote. adnah exists on the assumption that this is where the next generation of legal work will come from. So far, the evidence is holding.

Book a call