In the world of legal AI, where new tools frequently promise to revolutionize workflows, it can be easy to overlook the nuance of what lawyers actually need. Most don’t want to “disrupt” the profession. On the contrary, they just want to practice more effectively in a way that feels familiar to them.
This week, The Legal Wire sat down with someone who’s seen the space from all sides. Tanguy Chau, CEO of Paxton, spent a decade investing and building legal technology before founding Paxton. That background gives him a rare vantage point: not just how the legal AI market has grown, but where it’s fallen short.
TLW: How did your experience as an investor inform what you chose to prioritize (or avoid) when building Paxton?
Tanguy: “As a VC, I saw hundreds of startups pitch solutions to legal teams, and the best ones always had the same DNA: they solved a real pain point, fit seamlessly into workflows, and knew exactly who they were serving. That’s why Ironclad and Casetext stood out when I invested in them; they were precise, practical, and indispensable.
So when I started Paxton, I knew we had to take the same approach. We built with legal expertise from day one and focused obsessively on what matters most to lawyers: accuracy and trust. While others rushed half-baked tools to market, we invested in citation validation and reliability.
The investor in me knows sustainable businesses don’t win by cutting corners, and the legal tech founder in me knows trust is non-negotiable. Paxton exists at that intersection.”
A Platform That Covers the Essentials
Paxton is positioned as an all-in-one AI assistant, with capabilities across three fundamental pillars of legal practice: research, drafting, and document review. These functions aren’t new to legal tech, but Paxton’s core advantage lies in how it integrates them, and that it does so without overwhelming the user with complexity.
The platform’s research capabilities rely on a broad, regularly updated legal knowledge base, designed to return relevant case law and statutes with contextual awareness. Its drafting module, meanwhile, helps lawyers start and structure complex legal documents, not by generating walls of generic text, but by offering editable, authority-linked starting points.
TLW: In a space where many tools offer “assistive” drafting, what makes Paxton’s drafting engine more than just a template machine?
Tanguy: “Most tools are glorified template machines—sophisticated mad-libs that drop text into pre-set forms without understanding the law behind it. Paxton is different. Our drafting engine has legal reasoning built in. It doesn’t just generate words—it analyzes arguments, suggests relevant case law, anticipates counterpoints, and keeps every citation trackable and verifiable.
The result isn’t boilerplate; it’s work product that reflects real legal strategy. We built Paxton so attorneys don’t need separate tools for drafting and research—it’s one seamless workflow. The goal isn’t to replace judgment, but to amplify it. Paxton acts like a true drafting partner that understands both the law and the craft of persuasion.”
Human-Centric, Not Just AI-Powered
Paxton’s interface is conversational by design. The company’s tagline “just ask Paxton” isn’t marketing flair; it reflects a belief that AI should function less like software, and more like a teammate.
That idea becomes especially meaningful in the platform’s “thought partnership” model. Paxton doesn’t just answer queries. It proposes follow-up questions, flags overlooked issues, and adds context to support legal reasoning. In other words, it’s not there to replace human judgment, but to sharpen it.
TLW: How do you strike the balance between empowering lawyers and maintaining their autonomy, particularly in a system that’s designed to offer suggestions?
Tanguy: “The key to maintaining lawyer autonomy in an AI system is designing it as a research partner, not a decision-maker. Paxton never tells lawyers what to do; it provides comprehensive information, identifies relevant precedents, and offers drafting suggestions, but every strategic decision remains entirely in the lawyer’s hands. We’ve deliberately built transparency into every interaction: When it drafts language, it’s creating a starting point for your analysis, not a final product.
We’ve also prioritized auditability, every citation is traceable, and every output is verifiable. In analysis tasks, we provide sentence level highlighting and page level references, so lawyers can instantly see where each fact comes from and check the original context.
This approach actually enhances autonomy because it eliminates the time lawyers spend on mechanical research tasks and gives them more capacity to focus on the strategic thinking, client counseling, and judgment calls that define excellent legal practice. The most empowered lawyer isn’t one working without tools; it’s one who has access to comprehensive information quickly and can spend their time on the work that truly requires their expertise and training. We measure success not by how often lawyers accept our suggestions, but by how much time we free up for them to do their highest-value work and produce better outcomes for their clients, while maintaining complete control over every legal decision.”
Building for Scale Without Compromising Control
As more firms explore generative tools, concerns about security and data privacy have grown. Paxton’s approach to security caught our eye. It’s methodical and transparent. The platform adheres to SOC 2, ISO 27001, and HIPAA standards, and its internal policies extend from encryption protocols to access control reviews and vendor risk management.
While these credentials are increasingly expected, they’re still not standard across the board, especially among younger startups. For Paxton, however, compliance is central to enabling legal professionals to adopt AI with confidence.
TLW: How does your security posture affect the types of clients you’re able to serve, for example, particularly those in highly regulated fields like healthcare or finance?
Tanguy: “Security isn’t just a feature for us, it’s foundational to our product. Our customers work with sensitive information every day, on every matter, and we take that responsibility very seriously. Right from the day Paxton was founded, we understood that enterprise-grade security was absolute table stakes. So not only did we invest in building this secure foundation at the beginning, but we have continued to invest in this over time. Today, we’re SOC2 Type 2, ISO 27001, and HIPAA compliant, and have a BAA in place with all our providers.
But beyond technical safeguards, we have started to shift the conversation from: instead of asking “Can we use AI?” to “How can we use AI to better serve our clients while upholding compliance standards?” Most firms are realizing they need to evolve with the technology, but they are understandably careful to maintain their standards, and we can work with those standards. Firms with practices that touch highly regulated industries, such as healthcare or financial services, need to have a high standard for security and compliance. But, they are also managing complex use cases that could really benefit from technology.
We recognize that the most regulated industries often have the most to gain from AI assistance, only if the technology meets their stringent security requirements. Our enterprise-grade security capabilities don’t limit our market, they enable us to include practice areas or customers that handle the most sensitive documents and legal work.”
From Blank Pages to Better Practice
One of Paxton’s most pragmatic features is its ability to “jumpstart” drafting, which is a pain point many lawyers cite as both time-consuming and mentally taxing. With Paxton, users can initiate a legal document by describing its purpose or uploading a relevant source file. The platform then constructs a first draft with citations, structure, and edit-ready content.
This avoids the generative overreach of other platforms while still saving time, a delicate but important distinction.
TLW: What’s the philosophy behind the way Paxton structures its AI output, especially when it comes to citation transparency and editable prose?
Tanguy: “At Paxton, we believe legal AI should be a transparent partner, not a black box. That philosophy shapes everything about how we structure AI output.
First, you mentioned citation transparency. We understand how important it is to let users review the results directly. That’s why Paxton always includes inline citations linked right to the source material, whether it’s case law, statutes, or regulations, including highlighting the relevant sections within sources so it’s easy to validate.
Put simply, we don’t just show the answer; we also show where it comes from. We explicitly instruct our AI to use only factual information from our legal sources, and we force it to cite any piece of text it’s pulling from.
You can click to verify or open the whole case in context, so you’re never relying on hallucinated or unverifiable claims. This focus on transparency helps lawyers trust and validate the AI’s reasoning before incorporating it into client work.
Additionally, we know lawyers don’t want a robotic dump of text or to sift through a long back-and-forth chat; they want a usable output. We built Paxton to generate structured documents that follow legal conventions, or even their own templates if they prefer. So, whether it’s a demand letter, a motion to dismiss, or a factual summary, the output is clear and legible.
Everything a user generates in Paxton opens in our built-in editor, so you can revise, format, and finalize documents without switching between tools or windows. It’s about giving our users speed and control to research, draft, and iterate all in one platform.”
More Than a Product: A Strategy for Practice Growth
Tanguy’s vision for Paxton isn’t just about features. It’s about enabling law firms to scale, but not by adding headcount. Rather, Paxton focuses on boosting the output and confidence of teams that firms already have. In that way, Paxton isn’t competing with human talent. It’s extending it.
TLW: What have you learned from your early users about how Paxton impacts firm growth, workload distribution, or client satisfaction?
Tanguy: “We have heard so much from our users on how Paxton has helped their practices. Most common is the ability to actually grow their revenue. Some of our customers have even started to change their business models because of Paxton. One firm told us they’re shifting to monthly fee structures instead of hourly billing because of Paxton. Our product has helped them deliver their core services so much faster, they increased the number of cases they handle each month.
But, it’s not only about a more efficient operation for the practice. We’ve seen this benefit extend to their clients as well. One Personal Injury lawyer told us that Paxton surfaced specific details buried in the footnotes of the medical records that tripled the final settlement. When your AI can spot details that may be deep in the footnotes and may get missed by manual reviews, that’s not just efficiency, that’s real bottom-line impact.
We’ve also heard that Paxton is reshaping workload distribution for both attorneys and their extended teams. For example, we’re seeing paralegals contributing to attorney files “more effectively and with greater speed”. We have heard paralegals, legal assistants, or case managers can be more efficient with tools like Paxton, especially with some of the largest bottlenecks we see in the normal lifecycle of a case, for example, reviewing all the case facts and documents.
One quote that has stuck with me: “Paxton is my main research tool at this point. Just because it’s so much faster.” When your primary research workflow changes, everything downstream gets faster. Another lawyer mentioned they can now generate demand letters with just “minor tweaking” because Paxton handles the heavy lifting upfront.
Lastly, what we pride ourselves on is actually enabling our customers to serve their clients better and faster. The quality improvements, and ultimately better outcomes, are what really matter. We enable our customers and users to really operate at another level of efficiency and efficacy. One of our customers, Ferrigno Law, uses Paxton as a quality safeguard for final review to validate comprehensiveness or potentially catch oversights that could hurt client outcomes.
The pattern we’re seeing is firms using Paxton are able to punch above their weight, whether it be by taking on more complex cases, serving clients faster, or catching details that traditional workflows might miss. That’s the triple impact we designed for.”
