When legal professionals validate authority, the first thing they often notice is the citator indicators. An indicator provides an immediate visual cue about how a case has been treated by later courts...
For small and mid-size law firms, the calculus around expert witness research often feels straightforward: why pay an outside service when you have capable associates and paralegals who can run the searches...
Technology is only part of the equation. Successful AI adoption requires a partner who understands legal practice, firm workflows, and the realities of change management. When legal AI first entered...
Selecting outside counsel is one of the most important decisions an in-house legal department makes. The right law firm can influence litigation outcomes, costs, and overall risk exposure, making outside...
By Erica Frisby | Marketing Manager, LexisNexis® Legal & Professional At the LexisNexis® AI Symposium panel, “Lawyering in the Age of AI: Preparing the Next Generation for Practice...
* The views expressed in externally authored materials linked or published on this site do not necessarily reflect the views of LexisNexis Legal & Professional.
By Erica Frisby | Marketing Manager, LexisNexis® Legal & Professional
For years, the legal AI conversation focused on what lawyers could ask a tool to do. The next phase is about what AI systems can execute across a workflow.
That shift changes what legal leaders need to decide. AI is no longer only about helping individual lawyers work faster. It is becoming a question of how legal work should be organized, delivered, staffed, and measured.
At the 2026 LexisNexis AI Symposium, The Legal AI Inflection Point: Designing the Future of Legal Practice, legal and business leaders explored how agentic AI is changing the delivery of legal services. One panel, “Redesigning Legal Practice for the Client-Centric Agentic AI Era,” brought together perspectives from law firms, in-house legal teams, and legal innovation leaders.
The session was moderated by Roland Vogl, Executive Director and Co-founder of CodeX, The Stanford Center for Legal Informatics. Panelists included Danielle Benecke of Baker McKenzie, Kurt Chauviere of Blackstone, and Ilona Logvinova of Herbert Smith Freehills Kramer.
For legal leaders, the discussion pointed to a sharper question: not whether major law firms and legal departments will adopt AI, but what they will redesign because of it.
The first wave of generative AI in legal centered on interaction. Lawyers could ask questions, summarize documents, draft text, and test ideas through chat-based tools. Those capabilities remain useful, but agentic AI changes the frame.
Agentic AI systems can execute complex, multi-step tasks and operate within the workflows, harnesses, and organizational infrastructure around legal work. That changes the management question from “Which tools should we buy?” to “Which parts of the work should be redesigned?”
“The question now is whether stakeholders will use AI to retrofit what they’re doing or whether they will really rebuild from the ground up.” Roland Vogl, Executive Director and Co-founder, CodeX, The Stanford Center for Legal Informatics
Vogl framed the strategic choice as a choice between retrofitting and rebuilding. Retrofitting means plugging AI into existing structures, including traditional staffing models, review processes, and matter workflows. Rebuilding means asking whether those structures are still the right ones for delivering legal services in an agentic AI environment.
Benecke added another layer: AI should be treated as an organizational design issue, not simply a tool-selection issue. Many firms and departments will have access to similar models and applications. Differentiation will come from how they structure context, workflows, governance, and expertise around those tools.
For law firm leaders, that is a critical distinction. A tool rollout may improve individual productivity. A redesigned workflow can change the time, cost, and staffing assumptions behind pieces of legal work.
That difference becomes clearer when applied to real legal work.
Benecke described how an AI-native approach can change the shape of a large-scale global regulatory audit. Instead of relying primarily on manual review and analysis across jurisdictions, expert lawyers can conduct targeted interviews with business leaders, generate transcripts, and use AI-enabled workflows to analyze those transcripts against roughly 1,500 compliance points.
Human expertise is still central, but it is applied differently. Lawyers focus on judgment, context, interviews, and interpretation. AI helps compress the analysis timeline and broaden the scope of review.
Benecke also pointed to cybersecurity incident response, where facts change quickly, and reporting obligations may span regulators, consumers, employees, and multiple jurisdictions. In that environment, agentic workflows can help analyze evolving facts, forensic outputs, and legal requirements to identify preliminary notification obligations. That does not remove the need for legal judgment. It changes how quickly that judgment can be informed.
These examples move the conversation beyond efficiency. AI is not only helping legal teams do familiar tasks faster. It is opening the door to new service designs that would have been difficult, costly, or too slow under traditional delivery models.
Clients are looking past AI adoption
That shift matters because sophisticated clients are beginning to ask more precise questions about value.
From the client side, Chauviere described a pragmatic view of the market. Blackstone is less concerned with whether work is performed in-house, by a traditional law firm, or by an AI-native provider. The core question is: where does the company find the strongest combination of quality, speed, judgment, and value?
At Blackstone, Chauviere described a more granular approach to evaluating legal spending. Rather than focusing solely on total matter costs, his team is analyzing the specific tasks within those matters to understand where time and money are actually being spent. That level of visibility can help identify which parts of a matter require senior legal judgment, which parts may be suitable for automation, and how AI-enabled delivery should affect cost, timing, and staffing.
That scrutiny is beginning to show up in outside counsel conversations. Chauviere noted Blackstone expects AI-use disclosures to become part of panel-firm requirements, but simple disclosure is not the end goal. A notation that a tool was used does not, in itself, show whether the work became faster, better, or more cost-effective. The more useful conversation is whether AI has changed the time to deliver, the price point, or the way a specific piece of work is performed.
For outside counsel, the message is not that expertise matters less. It is that expertise must be easier to see, measure, and apply where it creates the most value.
Human judgment becomes the design constraint
As AI systems take on more responsibility, the limiting factor shifts from output to oversight.
Logvinova pointed to one of the most important challenges of agentic AI: human cognition becomes the bottleneck. AI systems can run multiple complex tasks at once. Humans cannot deeply supervise unlimited streams of work. That creates both an orchestration opportunity and an orchestration challenge: legal organizations need to decide how human intelligence and artificial intelligence should work together.
That means legal organizations need to design for judgment, not just output. Where should human review occur? Which lawyers are best positioned to steer AI systems? What information should the system receive? How should outputs be validated? When should risk be retained, and when should it be transferred to outside counsel?
Benecke captured that shift through the lens of expertise and accountability. In one client conversation, she described the decision to keep work in-house or send it outside turned on two questions: whether the legal department had the right expert to steer the AI system, and whether the organization was comfortable holding the risk. The tools themselves were not the differentiator. Judgment, accountability, and the ability to stand behind the work were.
In this environment, expertise is not replaced. It is repositioned. The lawyer’s role increasingly includes framing the problem, supplying the right context, supervising AI-generated work, identifying gaps, and standing behind the final judgment.
Access to models will not be enough. The competitive advantage will sit in the surrounding system: context, data, workflow design, governance, accountability, and client collaboration.
Benecke described this as a shift from prompt engineering to context engineering, and now toward “harness engineering.” A harness can range from simple instructions and structured files for an agent to a sophisticated legal technology environment. The strategic question is who owns that harness and where the intelligence around it compounds.
For law firms, this moves AI strategy closer to the center of the business. It now touches the core economics of practice: leverage, pricing, matter management, knowledge capture, professional development, and client value.
It also changes the conversation between firms and clients. Benecke described the opportunity for providers to work more closely with clients to identify major categories of work that can be accelerated, reduce headline costs over time, and free up capacity for more strategic, higher-value legal work. That kind of partnership is especially important as general counsel face pressure from the business to spend less on legal, while also managing new risks created by enterprise AI adoption.
The firms and departments that make the most progress will not necessarily be the ones that adopt the most tools. They will be the ones that understand where legal work slows down, where judgment truly matters, where risk should sit, and where clients are asking for a different model.
The practical question for leaders is no longer “How do we use AI?” It is “What should legal service delivery look like now that AI can execute more of the work?”
Explore how LexisNexis Protégé helps legal professionals use trusted legal AI to move from insight to action across modern legal workflows.