Legal AI Statistics: What Surveys Actually Measure

There is no single legal AI adoption rate, because the major surveys count different things. Wolters Kluwer's 2026 Future Ready Lawyer survey says 92% of lawyers use at least one AI tool in their daily workflow. Clio's 2025 Legal Trends Report says 79% of legal professionals use AI. 8am's 2026 Legal Industry Report says 69% personally use general-purpose generative AI at work. Thomson Reuters' 2026 AI in Professional Services Report says 41% of law firms are using generative AI as an organization. The ABA's 2024 Legal Technology Survey put lawyer AI use at 30%. All five can be true at once. The spread comes from three variables: whether the survey counts a person or an organization, whether "AI" means any tool or generative AI specifically, and who was sampled. Pick the number that matches the claim you are making.

We run a directory of legal AI software, we don't sell any of the tools in it, and featured listings are labeled and never change rankings. That makes adoption statistics something we read closely and have no reason to inflate. Every figure below names its source, and every source was opened and checked on 22 September 2026. Where we could only reach a publisher's summary rather than the full report, we say so.

Why do legal AI adoption numbers disagree so much?

Because "adoption" is not one question. Read the fine print on any headline figure and it resolves into one of four different measurements:

Individual use of any AI. Has this person used an AI tool for work at all? This produces the biggest numbers, because it catches a lawyer who asked ChatGPT to tidy one email.
Individual use of generative AI specifically. Narrower, and closer to what most people mean in 2026.
Organization-wide use. Has the firm or legal department adopted generative AI as an institution? This is much lower, because one enthusiastic associate doesn't make a firm an adopter.
Legal-specific tools. Is the person using a purpose-built legal product rather than a general assistant? Lower again.

Sampling is the second variable. A survey of a practice management vendor's own customers skews toward firms already comfortable with software. A global survey that mixes lawyers with tax, risk and government professionals reports blended totals unless you dig out the legal rows. A survey spanning the US, China and nine European countries isn't measuring the US market.

Timing is the third. The ABA figure most often quoted comes from a survey released in March 2025, and generative AI use moved sharply in the eighteen months after that. Quoting it as a 2026 number undersells current use. Quoting the 92% figure as "US law firms have adopted AI" oversells it.

What does each major survey actually measure?

Here is the reconciliation. "What it counts" describes the measurement as each publisher reported it; it is not verbatim questionnaire wording, which most publishers don't release.

Survey	Publisher	Sample (as reported)	What it counts	Headline number
2026 Future Ready Lawyer	Wolters Kluwer	810 lawyers in law firms and corporate legal departments across the US, China and nine European countries	Individuals using at least one AI tool in their daily workflow	92%
2025 Legal Trends Report	Clio	Surveys of 1,700+ US legal professionals (plus 1,000 consumers) and usage data from Clio customers	Legal professionals who use AI	79% (same as 2024)
2026 Legal Industry Report	8am	1,300+ US legal professionals, surveyed September to October 2025	Individuals personally using general-purpose generative AI at work	69% (up from 31% a year earlier)
2026 AI in Professional Services Report	Thomson Reuters Institute	1,500+ professionals across 27 countries in legal, tax, risk and government	Organizations whose teams use generative AI	41% of law firms, 47% of corporate legal departments (up from 28% and 23%)
2024 Legal Technology Survey Report	American Bar Association	US lawyers, results released March 2025	Responding lawyers using AI	30%

Sources for each row: the Wolters Kluwer 2026 Future Ready Lawyer announcement (10 March 2026); the Illinois Supreme Court Commission on Professionalism's summary of Clio's 2025 Legal Trends Report (29 October 2025), because Clio's own report pages refused automated access this run; 8am's 2026 Legal Industry Report page; the Thomson Reuters Institute's overview of the 2026 AI in Professional Services Report plus Thomson Reuters' legal blog, which breaks out the law firm and legal department rows; and the ABA Journal's report on the 2024 Legal Technology Survey (7 March 2025).

Which number to use for which claim:

"Most lawyers have tried AI at work": Clio's 79% or 8am's 69%. Both are US, both count individuals.
"Most lawyers now use generative AI tools like ChatGPT": 8am's 69%, which specifically counts general-purpose generative AI.
"Most law firms have adopted AI": none of these support it as stated. Thomson Reuters' organizational figure is 41% of law firms.
"AI is in lawyers' daily workflow globally": Wolters Kluwer's 92%, labeled as a multi-country survey of 810 lawyers.
"Adoption has roughly doubled in a year": Thomson Reuters (28% to 41% for law firms, 23% to 47% for legal departments) and 8am (31% to 69% for personal generative AI use) both show it.

One more trap. Clio's 2025 report found that only 40% of legal professionals use legal-specific AI, down from 58% in 2024, according to the same Commission summary. So growth in "AI use" is currently being driven by general assistants, not purpose-built legal software. If you're comparing legal tools rather than general assistants, our roundup of the best AI legal tools and the legal research category are the place to start.

How does adoption differ by firm size?

Larger firms report more AI use in every survey that splits by size, but the gap is narrowing.

ABA 2024 Legal Technology Survey: 46% of firms with 100 or more attorneys used AI, against 30% of firms with 10 to 49 lawyers and 18% of solo attorneys (ABA Journal).
Clio 2025 Legal Trends Report: 87% of legal professionals at large firms use AI, and 71% of solo firms still report using it (Commission summary).
Clio's 2026 solo and small firm report and its mid-sized firm report, as summarized by the North Carolina Bar Association's May 2026 roundup "By The Numbers: What Surveys Show About Law Firm AI Adoption": 71% of solo practitioners and 75% of small firms report AI adoption, and 86% of mid-sized firms say they have adopted AI.

The ABA and Clio figures aren't directly comparable (different years, different samples, and Clio's size bands differ from the ABA's), but the direction is consistent. The more useful finding for small firms is what happens after adoption, covered below. For a practical view of which tools fit a small practice, see our guide to AI for small law firms.

What are lawyers actually using AI for?

Drafting, research and summarizing, in that rough order, with legal-specific tools concentrated in research and document review.

8am 2026 Legal Industry Report (1,300+ US legal professionals), as reported in the ABA's Law Practice magazine coverage of the report: drafting correspondence 58%, general research 58%, brainstorming 54%, summarizing documents 47%. On time saved, 38% reported saving 1 to 5 hours per week and 14% reported saving 6 to 10 hours.

Thomson Reuters legal data (among legal professionals using AI tools, 2025 figures published on Thomson Reuters' legal blog in July 2026): document review 77%, legal research 74%, document summarization 74%, brief or memo drafting 59%, contract drafting 58%.

Wolters Kluwer 2026 Future Ready Lawyer: 62% of respondents report time savings of 6% to 20% of their work week.

ABA 2024 Legal Technology Survey: 52% of firms were using or considering ChatGPT, 26% Thomson Reuters CoCounsel and 24% Lexis+ AI.

The Thomson Reuters use-case numbers run higher than 8am's because they're calculated only among people already using AI, not across all respondents. That's a common source of confusion when two "use case" figures are quoted side by side.

Accuracy is the reason research tasks still need a human check. The Stanford RegLab study Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools found that Lexis+ AI, Westlaw AI-Assisted Research and Ask Practical Law AI each hallucinated between 17% and 33% of the time on its preregistered test set, fewer errors than general-purpose GPT-4 but far from zero. The ABA survey found three quarters of respondents cited hallucination concerns as the reason they had been hesitant to implement AI. The courts' response to that risk is tracked in our piece on AI hallucination sanctions. You can compare the research tools named here in the directory: CoCounsel and Lexis+ AI.

What do the surveys say about policies and ROI tracking?

This is where the gap between individual use and institutional readiness is widest.

Policies

8am 2026: 43% said their firm has no formal AI policy and no plans to create one. Only 9% said their firm has a written policy that is actually enforced. 54% said their firm has provided no training on responsible generative AI use and has no plans to (8am figures via the North Carolina Bar Association roundup and the ABA's Law Practice magazine).
Clio 2025: 53% of legal professionals said their firm has no AI policy or they are unaware of one. Among firms with a policy, 30% allow and encourage AI, 12% allow but don't encourage it, and 5% prohibit it (Commission summary).
Clio 2026 solo and small firm data (via the NCBA roundup): 57% of solos and 55% of small firms report no AI policy.

ROI and revenue

Thomson Reuters 2026 AI in Professional Services: only 18% of respondents say their organization tracks the ROI of AI tools, and another 40% don't know whether ROI is measured at all (Thomson Reuters Institute).
Clio 2025: 36% of legal professionals say AI has positively affected revenue, rising to 69% among those who have widely adopted it. A third said it was too early to tell.
Clio 2026 solo and small firm data (via NCBA): about a third report an AI-linked revenue increase (32% of solos, 31% of small firms), and 86% of solo firms and 78% of small firms have not adjusted pricing to account for AI.
8am 2026: only 6% of respondents said clients are explicitly pushing for AI-linked cost reductions.

Put those together and the pattern is clear. Individual lawyers adopted AI faster than their firms wrote rules for it, and most firms can't say whether it's paying off because they aren't measuring. For billing-hour firms, the Clio pricing figure matters most: time saved and not reinvested in client work behaves like an unplanned discount.

How to cite legal AI statistics without getting it wrong

A short checklist we use when reading any adoption figure, including in vendor decks:

Does it count individuals or organizations?
Does "AI" mean any AI tool, generative AI, or legal-specific software?
Who was sampled: US only, global, one vendor's customers, one firm-size band?
When was the fieldwork done, as opposed to when the report was published?
Is a use-case percentage calculated across all respondents or only among AI users?
Can you open the publisher's own page, or only a third party's summary of it?

If a statistic fails the last question, attribute it to the summary you actually read. That's the standard applied throughout this page.

Where this leaves tool buyers

The surveys agree on one practical point: most legal professionals are already using something, often a general assistant, often without a firm policy behind it. The decision in front of most firms is no longer whether to use AI but which tools to standardize on, and what can safely go into them. Browse the full legal AI directory to see which tools firms are choosing, by category and with verified pricing lines. If you build one of these tools and it's missing, submit it here; listings are free. For the broader question of what these tools can and can't do for a practice, see can AI replace a lawyer.

This page is industry analysis and general information about software, not legal advice.

FAQ

What percentage of lawyers use AI?

It depends on the definition. In the US, Clio's 2025 Legal Trends Report found 79% of legal professionals use AI, and 8am's 2026 Legal Industry Report found 69% personally use general-purpose generative AI at work. Globally, Wolters Kluwer's 2026 Future Ready Lawyer survey found 92% use at least one AI tool daily. At the organization level, Thomson Reuters reports 41% of law firms use generative AI.

Do small firms use AI more than big firms?

No. Every survey that splits by size shows larger firms ahead. The ABA's 2024 survey found 46% of 100+ attorney firms using AI against 18% of solos, and Clio's 2025 report found 87% at large firms against 71% at solo firms. Clio's 2026 data puts small firms at 75% and mid-sized firms at 86%, so the gap is narrowing.

What is the most common legal AI use case?

Among all respondents in 8am's 2026 survey, drafting correspondence and general research tie at 58%. Among legal professionals already using AI, Thomson Reuters' 2025 figures put document review first at 77%, followed by legal research and document summarization at 74% each.

Are law firms measuring AI ROI?

Mostly not. Thomson Reuters' 2026 AI in Professional Services Report found only 18% of respondents say their organization tracks AI ROI, and 40% don't know whether it's measured. Clio's 2025 report found 36% of legal professionals see a positive revenue impact, with a third saying it's too early to tell.