From investment-curious to investment-ready: how a litigation funder reshaped case prep around Lexity
Ankita MehtaIn our In Practice series we sit down with the people who actually use Lexity day-to-day. Not to demo features, but to ask what changed in their work, what they trust, and what's still missing. Episode five is the first interview where the user isn't a lawyer.
Peter Petyt is the CEO of Four Rivers, a litigation-finance advisor, and part of the wider GLF Funding group. His job is to take a claim, usually an international arbitration or complex commercial dispute, and prepare it to a standard a funder will read. We wanted his perspective specifically because the audience for AI in legal isn't only inside law firms. It is, increasingly, the whole funding ecosystem around them.
Executive summary
- The 20-page funding note that Four Rivers sends to funders, covering merits, defences, damages, budget, jurisdiction, enforcement, used to take "many days" of two-way work between lawyers, damages experts, and asset tracers. Today, with individual Clickflows™ stitched into the IC report, the same prep takes hours.
- The single biggest unlock isn't speed alone, it's the chronology. "Lexity performs this task admirably and very quickly. In minutes." The chronology is the document that lets a time-poor funder walk into the matter cold.
- The trust posture is propagated, not bilateral. GLF's own diligence on Lexity is something Peter actively communicates to every funder and every law firm he works with, confidentiality is, in his words, "one of the biggest talking points in AI in the legal sector at the moment."
- AI changes the origination market, not only the back office. The middle of the funding triangle, cases that aren't gold-plated but have potential with attention, is now economically workable, where it wasn't before.
- The boundary is honest: Lexity is "turbocharging the process at the start." The judgment, the tone, the funder-specific framing remain his.
Meet Peter
Peter is the CEO of Four Rivers, a litigation-finance advisor, and part of the wider GLF Funding group, a network of advisors worldwide who collaborate on origination, share know-how, and pursue funded cases together. The day-to-day job is to take a claim from a claimant or law firm, evaluate it, and prepare it for funders. The work spans international arbitration, complex commercial disputes, and high-value insolvency claims.
Peter's background is financial, which is why a piece of the work he focuses on most, case economics, damages valuation, budget benchmarking, sits in territory where a finance brain matters. The funding note he produces is a structured summary that has to do justice to all of that:
We like to produce what I call a funding note. The idea of that document is that within 45 minutes a funder can read it and understand everything they need to initially about a case, where the merits are, where the economics are, the risks. It's much longer than an executive summary. Essentially a 20-page note.
That document is the spine of the practice. Everything in this interview is, in one way or another, about how that spine got faster, cleaner, and more economically viable to produce.
First reaction: the careful sceptic
We asked Peter for his honest first reaction to AI tools coming into the legal market. He gave us the answer of someone who has seen new technologies introduced into other industries and watched the mistakes:
I've been in various industries, and certainly a couple where new technologies have been brought in and there were very large learning curves and a lot of mistakes, some quite damaging. So my initial reaction on any new technology is one of some scepticism. However, in legal technology we're seeing a really interesting and fast-moving dynamic. It's quite a fragmented, quite a difficult market to navigate.
He had used the obvious tools, Copilot, Claude, at the front end, "to make ourselves more efficient." But the difference with a platform purpose-built for the work was, in his words, the integration:
Lexity came across as the most suitable, efficient, integrated platform for what we needed. It took the benefits of the other AI tools we'd used, put them into one platform, then drilled down into the individual areas that require so much attention in advising on litigation funding. It enabled us to analyse and critique cases much more quickly at the front end.
Before Lexity: many days, many counterparties, many documents
We asked Peter to walk us through what a specific case used to look like before AI. He picked international arbitration:
You're gathering a huge amount of documents from a number of different sources. Usually starting with a detailed legal memo from the lawyers focused on the merits. But there's so much more required, the economics piece, damages, range of damages, different valuation methods. Plus the case budget. Plus asset tracing, collectibility, enforcement. These are crucial areas without which cases will not be funded.
The traditional process, review every document, assemble a digestible package for a funder, with a lot of two-way back-and-forth between lawyers, damages experts, and asset tracers, "could take many days."
That's the baseline. The interesting question is what changed.
The unlock: a chronology, in minutes
Asked what fundamentally changed about how he works, Peter went straight to one specific Clickflow™:
We've always produced a chronology of the case as part of our methodology. It's very important for anyone reading a case fresh to have a logical chronology of events, what happened when, who did what when. But you had to piece that together from multiple documents, and that was very time-consuming. Lexity performs this task admirably, and very quickly. If you feed in not just the lawyer's memo but a damages report, notes from calls, procedural orders from the tribunal, within a very short period of time, and I do mean minutes, Lexity is able to produce a detailed chronology of what happened and when.
This is the single most universal use of legal AI we encounter across the user base, sorting unstructured material into a defensible, cited chronology. It is also the document a time-poor funder, judge, or partner will read first. The cost of producing it used to gate the rest of the work. It no longer does.
Origination: working the middle of the triangle
The part of Peter's answer that we think is the most underrated is the one about the shape of the origination market. Most narratives about AI in legal stop at "saves time." Peter took it further:
The way I look at the market is a triangle. At the top, a handful of cases that are gold-plated and don't need our assistance. At the bottom, a whole load of cases that probably will never get funded, fatal flaws, merits or economics. We've been fairly good at analysing both. But there's a huge number of cases in the middle that have the potential to be funded but need attention. They need our value-add and know-how. This is where using Lexity really helps. We can feed things in, get a proper analysis early, decide this case has potential, here are the areas we need to work on. It filters out the ones that won't ever succeed and the ones that have potential.
In market-shape terms: AI lowers the unit cost of evaluating and improving a case. The number of workable cases, the ones it's economically sensible for an advisor to invest time in, expands. The middle of the triangle becomes a real market segment, not just a backlog. This is the kind of impact that doesn't show up in a productivity dashboard but does show up in the size of the business.
On trust: GLF diligence as a network primitive
We asked Peter how he thinks about uploading privileged case material into a platform: communications, strategy notes, financial terms. His answer is one we don't think people fully appreciate, because it points at how trust propagates in this market.
Within GLF we have a couple of people who are very savvy about this, they had a good look at Lexity, and were very happy with the security and confidentiality. That's a great comfort to me, and we communicate that to all the funders and law firms we deal with. It is one of the biggest talking points in AI in the legal sector at the moment, what is the confidentiality of data. I'm comfortable with the guarantees and reassurance Lexity has given us. And I think Lexity has done a lot of hard work to get themselves to a point where people can be comfortable with that.
Notice how this works. GLF, a sophisticated, well-resourced funder network, runs its own diligence on the platform. Peter then uses that diligence as a comfort signal, by name, in every funder conversation he has. The same diligence clears trust for the funders he speaks to, and for the claimant law firms he refers business with. One platform vetting becomes the implicit comfort statement for an entire network of stakeholders.
That is the part of trust in legal AI that is structurally different from B2B SaaS in other categories. It is not bilateral. It is ecosystem-shaped. And it is a major part of why a platform that earns the trust of the most cautious users ends up earning the trust of the rest of the chain by default.
There is a second reason this matters at the origination end of the work, not just the operational end:
Every meeting we have at the moment, we talk about Lexity. We talk about the AI capability we have. It's a definite advantage for us over some of the competitors we have in the market, that we are have that relationship, because people know and trust the methodologies and tools that we're using.
In this ecosystem, the AI partnership is itself a credentialling signal at the front of the sales conversation, not only in the workroom.
Live walkthrough: three Clickflows™ on a real test case
In a follow-up session, Peter walked us through how he runs Lexity on a typical case. The folder he set up, video test, held four anonymised documents that mirror the standard intake stack:
- An anonymised funding budget (the Excel spreadsheet showing capital deployment across the case lifecycle)
- An asset trace report from a third-party specialist
- An anonymised legal memorandum from the law firm, focused on merits
- An anonymised damages memo from a third-party valuation specialist
He then ran three Clickflows™ in parallel. The fact that you can run them in parallel is one of the small things that materially compounds across a case load:
The joy of Lexity is that you can run multiple Clickflows™ simultaneously.
The three he showed:
- Key Objectives & Overview of the Case. Press execute, point at the folder, submit. While that runs, move on.
- Master Case Chronology. "Very detailed, very quickly. The main events that have occurred in the case, the parties involved, a description, links back to the input documents. Really useful as to what happened when over a period of time." In his test case, the first major event was 2011 and the last 2025: fourteen years of file, in chronological order, in minutes.
- Case Merits Assessment, including quantum. "A summary of the merits of the claim. An anticipated defences section. And quantum, clearly from a funding point of view one of the most important aspects." The output included a $40-60M USD dispute valuation across different methodologies, with deep links back to the original documents.
He then opened Ask Docs to interrogate specific points and showed the in-platform translation across 30+ languages, relevant given the international nature of the case load.
The throughline:
A comprehensive, integrated suite of tools you can use to very quickly produce a pack you can send to funders.
We asked whether he could see himself going back to doing any of this manually:
No, not at all. This is like having a whole team of people behind you. Efficiency and speed are very important in this industry.
Where AI ends, judgment begins
We asked Peter the question we ask every interviewee in the series: where does the tool stop, and where do you start. His answer was, characteristically, both honest and concrete:
You do need to review the outputs. Nothing is perfect. Reviewing the outputs and adding your own personal opinions and judgment to the documents is still very important. I have a certain way of writing, a certain way of communicating. So sometimes I'm changing the tone of some of the content before I send a final to a funder. And maybe reorganising some of the content. But this is basically turbocharging the process at the start. What would have taken me days, if not weeks, to get to a point where I'd be reviewing and playing around with content, is now a matter of hours.
That is the right posture. The platform produces the structured first pass. The advisor takes it from "competent" to "this is the document you can put in front of an investment committee." The compression is at the front end, not at the judgment end.
Honest critique
We asked what he wants next. The answer is the part of the funding work that, in Peter's view, the market still hasn't built well enough: the economics.
With my financial hat on, I'm always interested in exploring improvements in the economics of cases, preparing budgets, extracting data, benchmarking against other cases we've done. Funding is a high-risk environment. If you can produce the same case for 50% less funding by being creative and innovative, the chances of it being funded are much improved. There's continued development required in those areas.
That feedback is on the development plan and we'd rather acknowledge it than dress it up. Budget benchmarking across a portfolio of past cases is exactly the kind of thing Lexity should be doing at maturity, and currently isn't doing as well as it should.
What he'd tell a colleague
Asked what he'd say to someone considering Lexity, Peter went short and ecosystem-shaped:
If all stakeholders in the case have the ability to work with us using the Lexity platform, to see the results, that can only be a good thing.
That's the funder's version of the recommendation. It's a recommendation about a medium, a shared way for advisors, law firms, claimants, and funders to interact with the same case material at the same level of structure. We agree with him. That's what we are trying to be.
What changed about the work itself
We closed the conversation with the same question we ask every interviewee: what fundamentally changed about the practice, in one word.
Efficiency. And speed. Both, much improved as a result of our use of Lexity.
We'll add the word the rest of the interview points to: capacity. Peter and the GLF group can now work the part of the funding triangle they couldn't economically work before. That isn't a productivity story. It's a scope story. And it's the kind of change AI in legal will increasingly be measured against.
