Desiree Sainthrope is a leading voice at the intersection of global compliance and emerging legal technologies. With a career spanning the complexities of international trade agreements and intellectual property, she offers a unique vantage point on how artificial intelligence is currently restructuring the very foundation of legal practice. As “hybrid” law firms emerge—entities that blend traditional counsel with sophisticated software development—she explains why this moment represents a fundamental shift in how legal services are delivered and regulated.
U.S. regulations often require hybrid law firms to maintain a separate tech entity to secure outside investment. How does this structural split impact the daily collaboration between software developers and attorneys, and what specific steps are taken to ensure these distinct operations deliver a seamless experience for the client?
The structural divide between a law firm and its tech-funded affiliate is a delicate balancing act that requires constant communication to prevent the “two-island” problem. For instance, a firm like Covenant must maintain Covenant Law for its legal practice while Covenant Inc. serves as the startup arm that attracts investor capital. In daily practice, this means developers aren’t just building tools in a vacuum; they are often working alongside practitioners to ensure that software like an MFN election tool actually solves the granular headaches lawyers face. To keep the client experience seamless, these firms often use a unified front-end where the client doesn’t feel the friction of moving between a tech platform and a legal consultation. It is a rigorous process of ensuring that while the money and the corporate structures are separate to satisfy regulators, the workflow feels like a single, high-speed engine of efficiency.
Modern legal models are increasingly placing artificial intelligence at the core of service delivery, with humans serving primarily as a governance or judgment layer. How does this shift redefine the traditional billable hour, and what specific anecdotes can you share regarding how human oversight has successfully caught errors in automated outputs?
We are witnessing a fundamental redesign of the legal service delivery model where technology is no longer an accessory, but the primary engine. This shift moves the human attorney into a “governance layer,” where their value is tied to high-level judgment rather than the sheer volume of hours spent on manual document review. While the billable hour has long been the industry standard, AI-native firms are finding that clients prefer results-oriented pricing because an AI can process what used to take twenty hours in a matter of seconds. However, the human element remains vital; for example, an attorney supervising an AI agent might notice a nuance in a complex regulatory filing that the model misinterpreted due to a lack of local context. It is in those moments of “judgment-based oversight” where the lawyer prevents a costly hallucination or error, proving that the human touch is the ultimate safeguard in an automated world.
While earlier attempts to blend legal services with software development struggled to scale, today’s landscape is defined by vastly improved model capabilities. What specific technological advancements now exist that were absent during previous market cycles, and how do these tools solve the integration issues that led to past failures?
The most glaring difference between the failed efforts of 2017 and today’s success stories is the emergence of generative AI and Large Language Models. When Atrium launched in 2017 and eventually raised $75.5 million, the underlying technology simply wasn’t sophisticated enough to handle the linguistic nuances of law, leading to a disconnect between the software’s promise and its performance. Today, we have moved beyond simple “if-then” automation into a realm where models can truly understand and generate complex legal prose. These newer hybrid firms are better positioned because they aren’t just trying to force-fit old technology into a law firm; they are building on a foundation of “native” AI that is inherently more flexible. This leap in model capability solves the integration issues of the past by providing a technological core that is actually powerful enough to support a full-scale legal practice.
AI-native firms are now deploying specialized tools for tasks like NDA mark-ups, LPA reviews, and due diligence. What metrics are you seeing in terms of efficiency gains for these specific workflows, and could you provide a step-by-step breakdown of how a firm transitions from manual review to an AI-fronted process?
The efficiency gains in specialized workflows like NDA mark-ups and Limited Partnership Agreement (LPA) reviews are nothing short of transformative for the modern practitioner. To transition, a firm typically begins by feeding its historical data and preferred “playbooks” into the AI, teaching the model what a “good” mark-up looks like for that specific client’s risk appetite. Next, the firm implements an AI-fronted intake where the software performs the first pass, highlighting deviations and suggesting standardized clauses within seconds. The final step involves a human lawyer reviewing the “redlines” to ensure the AI hasn’t missed a subtle strategic pivot, which drastically reduces the time spent on rote tasks. This transition allows firms to handle a much higher volume of due diligence without increasing their headcount, effectively decoupling revenue from the number of bodies in the room.
Attorneys working within hybrid structures face unique pressures to satisfy state bar qualifications while operating alongside tech-funded affiliates. What are the most significant ethical risks when balancing investor interests with legal duties, and what protocols must be in place to protect a practitioner’s license in this experimental environment?
The primary ethical risk in a hybrid model is the potential for investor pressure to compromise a lawyer’s independent professional judgment. Because these tech entities receive outside funding, there is always a fear that the drive for “startup-style” growth might conflict with the slow, meticulous duties a lawyer owes to their client and the court. To protect their licenses, practitioners in states like Illinois or Michigan must maintain a “colorable” and robust wall between the tech company’s business goals and the law firm’s ethical obligations. This involves strict protocols on data sharing, ensuring that attorney-client privilege is never compromised by the tech side’s data analytics needs. If a lawyer cannot prove that they—and not the investors—are the ones making the final legal calls, they face the very real threat of a state bar grievance and the loss of their professional livelihood.
What is your forecast for hybrid law firms?
My forecast for hybrid law firms is one of rapid expansion followed by a period of intense regulatory scrutiny as they move from the fringes to the mainstream. We are currently in an “exciting moment” where the market improvement in AI models’ capabilities makes these firms feel more viable than the first wave of legal tech startups. I expect that within the next five years, the “hybrid” label will start to disappear because every competitive mid-to-large firm will essentially operate as a tech-enabled entity. However, the ultimate success of this model will depend on whether these firms can maintain the delicate balance between high-speed technological innovation and the rigid ethical standards that the legal profession demands. We will likely see a new set of bar rules specifically tailored to these structures to ensure that the “judgment layer” provided by humans remains the primary authority in the delivery of legal advice.
