By Kim Xi Harris |Founder & Platform Architect, Lex Arca™ Legal Vault | Calculate your firm’s billing leakage | legalvault@lex-arca.com

According to Clio’s 2026 Legal Trends Report for Solo and Small Law Firms (May 2026, https://www.clio.com/about/press/2026-solo-small-firm-report/), 71% of solo practitioners and 75% of small firms are now using AI to complete legal work — yet fewer than 33% have seen any revenue increase from it, compared to nearly 60% of enterprise firms. The gap between AI adoption and AI results is not a training problem. It is an architecture problem.

Supio’s 2026 report, presented at the Association of Justice Annual Convention, found that 73 percent of plaintiff attorneys believe AI can benefit their practice — but only 30 percent have actually embedded it in their work. The reason is not price. It is source traceability: plaintiff attorneys handling catastrophic injury and wrongful death cases cannot use AI output they cannot trace back to a primary source, because unverifiable research is a liability risk their clients cannot afford.

What Did the Supio Report Actually Find?

The Supio Report surveyed plaintiff attorneys at the Association of Justice 2026 Convention and found a 43-percentage-point gap between AI belief and AI adoption. Seventy-three percent of respondents said they believe AI can improve their practice. Thirty percent have embedded it in their actual workflow.

The dominant barrier was not cost, not technology complexity, and not lack of training. It was source traceability. Plaintiff attorneys building catastrophic injury, wrongful death, and mass tort cases need to know exactly where every fact came from — what document, what study, what deposition, what court record. Current mainstream legal AI tools synthesize answers from indexed sources without producing a verifiable link to the primary material. That synthesis is useful for analysis. It is a liability when opposing counsel asks for the source and the attorney cannot produce one.

The gap is not philosophical. Plaintiff attorneys understand what AI can do. They will not use it at scale until they can show that the research it produced is traceable to a document that can be produced in discovery, disclosed to a client, and defended at deposition.

Why Does Source Traceability Matter More for Plaintiff Attorneys Than for Other Practitioners?

Plaintiff attorneys carry a specific evidentiary burden that creates a distinct AI risk profile. In a catastrophic injury case, every expert opinion, every damages calculation, and every legal argument must be traceable to a source that will survive cross-examination and Daubert scrutiny. If an AI tool produces a medical study summary that the attorney relies on — and that summary turns out to misstate the study’s findings, or cite a study that was retracted, or conflate two separate publications — the consequence is not a sanctions order. It is a client’s case.

The 30 percent of plaintiff attorneys who have embedded AI are not more risk-tolerant than the 70 percent who have not. They have found tools or workflows where the output is traceable to a source they can independently verify — or they are using AI for tasks where the evidentiary stakes are lower. Brief drafting, internal research summaries, intake automation. Not the research that goes into the case record.

Understanding the ABA Opinion 512 compliance workflow is relevant here because it places the verification burden squarely on the attorney — not the platform. A plaintiff attorney who relies on unverifiable AI research has not outsourced the risk to the vendor. The attorney owns it, and so does the client.

What Architecture Closes the Plaintiff AI Trust Gap?

The source traceability problem is an architecture problem, not a training problem. General-purpose AI legal tools are built to generate answers. They are not built to produce the trace chain that shows a plaintiff attorney exactly which document generated each component of the output.

The architecture that closes the plaintiff trust gap has three requirements. First, AI-generated output must link back to the specific document — not the document category, not the indexed source, but the actual file the model referenced. Second, the attorney must be able to verify that link against the primary source before it enters the case record. Third, there must be an append-only, tamper-evident record of that verification — so that when opposing counsel asks how a fact was sourced, the answer is not “the AI found it” but rather a timestamped verification log showing the attorney confirmed it against a specific document.

A litigation intelligence platform for solo firms built on a local-first private vault addresses this architecture problem directly — keeping client files out of third-party cloud infrastructure while producing the traceable research record plaintiff attorneys actually need. The 43-percentage-point gap the Supio Report identified is the market that exists on the other side of solving that architecture problem.

From Kim’s Chair: The Questions I Would Have Asked

I did not build Lex Arca™ Legal Vault by studying plaintiff attorney workflows in the abstract. I built it because I understand what it means to be on the receiving end of a case outcome — to have trusted the research behind a filing and to have had no way to know whether that research was traceable, verified, or assembled by a tool no one in that firm had actually checked. When I read the Supio Report’s finding that 70 percent of plaintiff attorneys who believe AI can help their practice have not adopted it, I do not see a technology adoption curve. I see the clients those attorneys represent — whose cases deserve both the efficiency AI can provide and the verification standard those attorneys are right to demand.

If I were in the room at the Association of Justice 2026 Convention when the Supio Report was presented, here is what I would have asked:

1. The Report found a 43-percentage-point gap between AI belief and AI adoption among plaintiff attorneys — what specific feature or safeguard, if it existed in a legal AI platform today, would close that gap, and which platforms have built it?

2. Plaintiff attorneys handle cases where research errors do not produce sanctions orders — they produce adverse verdicts for injured clients. What accountability mechanism exists for a legal AI platform that provides unverifiable output that enters a plaintiff’s brief or an expert’s report?

3. The 30 percent of plaintiff attorneys who have adopted AI — are they using it on the research that goes into the case record, or on lower-stakes internal tasks where the traceability requirement is less acute?

4. If a plaintiff attorney at this convention uses an AI tool on a catastrophic injury matter and that tool’s output contains a fabricated study citation that goes undetected until cross-examination, what does the client’s recourse look like against that attorney — and against that platform?

And if I were your client — someone whose catastrophic injury case you handled with AI research assistance — here is what I would have asked you:

1. When AI surfaced a key fact about my injury that you relied on, can you show me exactly which document that fact came from — the original source, not the AI’s summary?

2. If the AI research you used on my case contained a fabricated or misattributed source, is there a verification record showing you confirmed it independently before it entered my file?

3. You told me AI helps you build stronger cases for clients like me — but the Supio Report found that 70 percent of plaintiff attorneys who share that belief haven’t adopted it because of source traceability concerns. What is the verification protocol that distinguishes your use from the practices those 70 percent declined to adopt?

4. If my case went to trial and opposing counsel challenged the source of a key fact from your AI-assisted research, what would you hand the court to show your verification process?

The 70 percent of plaintiff attorneys who have not adopted AI are not making a mistake. They are making a correct read of the risk. The question is whether the architecture to eliminate that risk exists yet — and whether the 30 percent who have adopted AI are using it in a way that their clients could defend.

Key Takeaways

1. Supio’s 2026 report presented at the Association of Justice Annual Convention found a 43-percentage-point gap between plaintiff attorney belief in AI (73%) and actual adoption (30%), driven primarily by the absence of traceable, source-linked AI output.

2. Plaintiff attorneys handling catastrophic injury, wrongful death, and mass tort cases face a higher traceability standard than other practitioners — AI output that cannot be linked back to a verifiable primary document is not a research tool; it is a liability.

3. Closing the plaintiff AI trust gap requires a platform architecture that produces traceable output and a documented, timestamped verification record — not training programs or general AI literacy initiatives.

4. Lex Arca™ Legal Vault provides a documented, verifiable AI activity trail designed to support attorney compliance workflows, including the source-traceable research record that plaintiff attorneys actually need before they can adopt AI at scale.

5. Calculate your firm’s billing leakage and get early access at https://calculator.lex-arca.com.

About the Author: Kim Xi Harris is the Founder and Platform Architect of Lex Arca™, an AI-native litigation intelligence and compliance platform for solo and small-firm attorneys. She is a Cornell Women’s Entrepreneur Program graduate, SBA Women in Business Champion Award recipient, WOSB certified, and holds five Google AI certifications. Calculate your firm’s billing leakage at https://calculator.lex-arca.com — or reach us at legalvault@lex-arca.com.