What will AI change in 2027?

This summer we published our State of AI Readiness in Legal 2026 report, based on a survey of 160 legal professionals across the US and UK. The headline was clear: AI adoption had raced ahead of the readiness needed to use it safely.

87% of respondents were using or experimenting with AI, but only 14.4% were very confident it delivered real value. 65.6% said their organisation wasn't ready, or wasn't sure it was ready, to scale AI safely. A client-facing mistake caused by AI was the single biggest fear, and 43% weren't confident they knew which AI tools their own lawyers were using day to day.

That was only a few months ago, but in AI terms it's a long time. New tools have launched, existing tools have added AI features, and firms that were watching from the sidelines have started rolling things out. Some of what felt true in the summer may already look different.

So before the year ends, we want to refresh our thinking, and we'd like your help.

A quick look ahead

We've put together a short survey asking how you see AI changing legal next year. It's three questions, plus one optional one about your role, and it takes about a minute.

We're asking:

• Where your team's time will go. Will knowledge and research teams spend more time on higher-value analysis, checking AI output, teaching people which tools to use, leading the firm's AI strategy, or will things look much the same?
• What will make the difference. Is it training, giving AI one clear owner, starting from the problems firms actually need to solve, fixing data and knowledge first, or moving fast on new tools?
• What will get in the way. Too many tools each with their own AI, output that looks right but isn't, keeping training up with the pace of change, losing the junior talent pipeline, or simply not knowing which tools people are using?

There are no right answers. We want your honest view of what's coming.

Why we're asking now

The end of the year is when most of us start planning for the next one. Budgets are being set, priorities agreed, and decisions made about which tools stay and which go. Knowing how your peers see the year ahead is useful context for those conversations, especially when the ground keeps shifting.

It also lets us test whether the picture from the summer still holds. Is the worry about AI output that looks right but isn't still top of the list? Do teams have a better handle on which tools are in use? Or have the challenges moved on altogether?

See how your view compares

Once the results are in, we'll share how the predictions stack up across knowledge, library and research teams. If you'd like to see how your answers compare with your peers, just leave your details at the end and we'll email you the results. That part is optional, so you can submit your predictions on their own if you prefer.

It's one minute of your time, and it'll give all of us a better sense of what next year really looks like.