Meredith-Flister: Personal project
A solicitor who knows how to code has created an AI adversary that stress-tests legal arguments before they are tried in court.
Larissa Meredith-Flister designed and built the Opposing Counsel Review to help lawyers challenge their own thinking so they can prepare stronger cases.
Lawyers feed their legal argument, draft submission, witness statement or structured reasonings into the AI.
The AI generates a series of qualifying questions, depending on the nature of the case.
For example, if a lawyer put a witness statement into the AI, the qualifying questions would relate to evidential gaps and inconsistencies. Feeding in a skeleton argument would generate questions about issues such as causation.
With the qualifying questions answered, Opposing Counsel then dismantles your legal argument, “attacking, undermining and exposing weaknesses” in seconds.
Opposition Counsel is available as a free AI skill via Lawvable [1] – meaning it teaches AI assistants like Claude, ChatGPT or Gemini how to perform a specific legal task.
Creating Opposing Counsel was a personal project, outside her work at London litigation boutique Charles Lyndon, where she is an associate solicitor in the competition team.
She is a member of networks like Legal Quants and RepresentAI, who she described as “phenomenal”, where lawyers and coders meet, share ideas and create new pieces of lawtech.
Ms Meredith-Flister told Legal Futures: “Most lawyers don’t stress-test their arguments. They refine, they polish, and they make things sound more persuasive, but they never try to break what they’ve built. Opposing Counsel forces you to confront your assumptions, your bias, and shows you where your reasoning is fragile.”
“It’s not a proofreader or a summariser – it’s an adversary, and that’s the point. The value isn’t in what it tells you is good about your work. It shows you exactly where your argument would break under hostile scrutiny, so you can fix it before it reaches the opposing side.”
When the stress-test is complete, Opposing Counsel sets out a structured adversarial analysis of the legal argument.
The AI identifies the strongest line of attack that the opposition could use to defeat your argument.
It tests the logic in the reasoning to expose any rhetoric or fragility in your argument, highlights evidential holes and inconsistencies, and shows how a sceptical judge would respond to the argument.
Finally, it provides three to five high-impact points to deliver in oral submissions.
Ms Meredith-Flister outlined how Opposing Counsel would stress-test a competition law submission, based on the argument that a dominant firm’s exclusive supply agreements with major retailers amounted to an abuse of dominant position, citing a 60% market share and the foreclosure of rival suppliers from key distribution channels.
She said: “The skill would identify that the submission never defined the relevant market and that a broader market definition would dilute dominance.
“It conflated the existence of exclusive agreements with anticompetitive effect, without proving rivals could not reach consumers through alternative channels. And it offered no economic evidence quantifying the alleged foreclosure.”
“The argument may read persuasively, but the skill exposes that the writer has assumed dominance and abuse from market share alone, without doing the analytical work that a competition tribunal would require.”
Ms Meredith-Flister combines legal expertise and coding skills. She said her work “sits at the intersection of UK competition litigation, data privacy and AI.”
“Alongside my litigation practice, I am interested in practical AI and legal workflows; how lawyers can use these tools in ways that improve efficiency without compromising accuracy, trust, or professional judgment.
“I build and think about systems for people who care not just whether an output is fast, but whether it is reliable, verifiable and safe to use in high-stakes work.”
The idea behind Opposing Counsel came from an insight that Ms Meredith-Flister had into how personal bias crept into a lawyer’s work and thinking.
She said this bias was part of human nature, and found in all professions, but with lawyers it could limit how they prepared their cases, even if they asked colleagues for opinions on a legal argument.
“Other people on the team tend to be polite, rather than challenge your thinking. AI won’t do that. It will just deal with the case, and it will augment what you are doing – adding to and supporting your human skills.”
The lawtech market is seeing more innovations coming from individual lawyers in their spare time.
Earlier this month, we reported on the launch of BunTool [2], an AI product that makes hassle-free court bundles built by Tris Sherliker, an IP lawyer at Bird & Bird.