Cleared the Budget for AI? That's Not the Hard Part

Given how eager businesses seem to use AI, budget was always unlikely to be the reason AI stalled.

Our State of AI Readiness in Legal 2026 report backs that up: 87% of legal professionals are already using or experimenting with AI. The money got approved. What’s actually holding firms back shows up afterward, in a set of compliance and governance questions nobody priced in when the budget was signed off.

It’s worth looking at what’s actually keeping legal leaders up at night.

Client-facing mistakes top their list of fears, at 58.8%. Regulatory breach isn’t far behind, at 44.4%. In practice, those two fears tend to be the same one wearing different clothes: in a regulated profession, an AI answer that sounds confident but is wrong doesn’t stay a private embarrassment for long. It becomes a compliance incident, whether or not anyone catches it right away.

A lot of firms are more exposed to that risk than they might realize - firms without a formal AI governance framework are roughly twice as likely to have no real visibility into which AI tools their own lawyers are actually using day to day (64% compared with 28% for firms that do have a framework in place). You can’t govern a tool you don’t know is being used. And you certainly can’t explain it to a regulator.

A policy on paper isn’t the same as being ready

Here’s the part that should give even the more prepared firms pause.

Knowledge and innovation leaders report having a formal AI governance framework at nearly double the rate of the broader legal market. Ask that same group whether they actually feel ready to scale AI safely, though, and the gap nearly disappears. Writing a policy is one job, enforcing it is another, and most firms have only done the first.

Emily Florio, Director of Knowledge Research and Resources at DLA Piper, put her finger on exactly how that gap plays out day to day. Speaking at our August 2026 online panel, she said: “It’s got citations now, so I must be okay.”

It’s an understandable instinct. But a citation only looks like verification. It isn’t one. It’s a format. And treating that format as proof is precisely the kind of confident mistake a good governance policy is meant to catch, assuming the firm has the structure behind it to catch anything at all.

What closes the gap

That underlying structure is exactly what our AI Readiness Diagnostic sets out to measure, across three foundations: structured knowledge, meaning can people actually find and trust what the firm already knows; clear governance, meaning is there a real framework and does anyone have visibility into what’s being used; and connected systems, meaning do the tools talk to each other or does everyone rely on remembering to check three different places. None of that comes bundled with a budget line, however generous.

So if your firm has already cleared the budget hurdle, the real question left is whether you’re actually ready to scale safely. Six questions and about a minute will tell you where you stand, click below.