At ILTACON this year, I met with Chris Haley, Vice President of Practice Empowerment aiR at Relativity, on the current and future impact of Artificial Intelligence, particularly Agentic AI, within the legal sector. Here are the highlights of that interview.
I. AI as a Transformational Force (Beyond Incremental Change)
Haley emphasized that AI represents a “transformational change to how we do work,” rather than just a minor software enhancement or workflow adjustment. He likens it to previous technological shifts but asserted this one will be the “biggest one yet,” fundamentally altering the legal industry and human endeavors. Relativity’s role is to guide clients and internal teams through this “journey of going from where we are today to AI and now AI agentic journey in the future.”
II. Understanding Agentic AI: Beyond Generative AI
A significant focus is placed on differentiating Agentic AI from Generative AI. Haley explains that if human intelligence in the Stone Age was about humans using tools, “the dawn of the AI agentic age is this tooling of AI, allowing it to take action more efficiently.”
- Generative AI: Primarily acts as an assistant, answering questions or drafting documents.
- Agentic AI: Builds on generative capabilities by enabling AI to “take action and work amongst different AI components and human components” to “tool that experience,” making work more efficient.
III. Practical Applications of Agentic AI: Orchestrated Workflows
Haley provided a concrete use case for Agentic AI in legal document review:
Instead of a single AI agent reviewing documents for specific elements, an orchestration agent can guide complex workflows.
Example: A user requests a review of documents “for a specific purpose,” and the orchestration agent sets up the entire workflow, including:
- Triggers for human intervention if errors occur or validation is needed.
- Specific AI agents to evaluate for privilege.
- Subsequent evaluation of results for personal information.
This approach eliminates “manual intervention in between each step of the process,” streamlining complex tasks and completing the work faster.
IV. Mitigating Hallucinations in Legal AI
Recognizing the significant concern around AI hallucinations and accuracy, particularly in legal filings, Haley outlined a three-pronged strategy:
- Technological Grounding: Implement features that enable products to “check themselves.” Relativity’s products, for instance, “have a specific feature that looks at the citations that that our product returns” and “checks to verify that those citations that it’s relying on are actually found in the text of the document.” This technical check ensures that the citations our aiR products rely on to make its predictions are not hallucinated
- Human-in-the-Loop Process: Emphasize the crucial role of human oversight. Relativity’s products include “validation tools” and encourage “sampling of the results to confirm the accuracy of those results,” providing a “reasonable step” statistically measuring accuracy to have confidence that the results are accurate.”
- Segmenting High-Risk Tasks: For creative or drafting tasks, which carry higher risk, break them into “smaller pieces.” Haley advises, “I don’t want to ask it to draft an entire research report for me… Instead, I might chunk that or break that up into smaller pieces or requests.” This approach mirrors how one would manage a junior team member: “Frankly, I would treat it like a human. I would do the same thing with any assistant or junior member of our team. I try not to ask them to complete all the work and only come back when you’re done. No, check in with me early and often so that I can see the progress and provide feedback before any work goes in the wrong direction.”
V. AI’s Impact on Legal Careers: Opportunity, Not Job Loss
Haley discussed fears of widespread job displacement, drawing parallels to past technological revolutions (e.g., digital video editing, the internet, email). He expresses “tremendous faith that those of us that lean in… and accept and work on it and learn and be curious like we have in all the technology transformations before this one will be successful.”
- Acceleration of High-Value Work: AI acts as an “accelerant to help me do the most important things, the higher value work, the things that I never got to do because I was too busy with the tedious routine and mundane stuff.”
- Creation of New Opportunities: Rather than taking jobs, AI will “create more opportunity” by freeing up time for problem-solving and creative endeavors currently ignored due to time constraints.
- Mental Retooling Required: Professionals will need to “mentally retool” and rethink prioritization, discipline, and habits to leverage this newfound capacity.
VI. Billing Practices: The Shift Towards Value and Task-Based Models
Haley acknowledged the ongoing debate around billable hours versus task-based billing, especially with AI driving efficiency.
- He advocated for “tying the cost or what you charge for something to its value,” preferring task-based billing as a “good behavior modifier.”
- Despite IDC study indicating a significant portion of professionals see more task-based billing, Haley said he is “not convinced that we’re going to suddenly see some acceleration to task based billing take over from hourly billing,” citing the slow pace of change within the legal industry.
- The goal is for billing to reflect “the insight we obtained regardless of the time spent that led to a better outcome,” rather than simply the volume of work performed.
VII. Accelerating AI Adoption and Future Outlook
Relativity has observed a “large adoption curve” for its AI products like “air,” with discussions now shifting to “scale and performance” rather than just accuracy.
- Faster Adoption than Previous Technologies: “Much faster adoption rate than CAL ever was,” attributed to the diverse use cases aiR products can handle, beyond the narrow focus of CAL and classic TAR (Technology Assisted Review).
- Widespread Use Cases: Adoption is strong in areas like DSARs (Data Subject Access Requests), privacy issues, and creative workflows, where generative AI can “comprehensively analyze documents and return a result.”
- Future Developments: Haley expresses unprecedented excitement for the “next 6 to 12 months,” teasing that they “have just scratched the surface at Relativity for what they think can be done in the future.”
Conclusion
Chris Haley painted a picture of AI as an indispensable and rapidly evolving force within the legal industry. From its foundational shift in how work is done to the emergence of advanced Agentic AI, the message is clear: AI is not merely a tool but a partner in complex legal workflows. While challenges like hallucinations require robust mitigation strategies, the overall outlook is one of immense opportunity for legal professionals to elevate their work, drive greater value for clients, and adapt to a future where AI empowers rather than replaces human expertise. The rapid adoption rate of AI tools, particularly in diverse applications, signals a significant paradigm shift currently underway.
So, what do you think? How much has AI transformed the legal industry in recent years? Please share any comments you might have or if you’d like to know more about a particular topic.
Note: This write-up from a recorded interview was prepared with the assistance of NotebookLM from Google and reviewed and edited by both Chris and me for accuracy.
Disclaimer: The views represented herein are exclusively the views of the author, and do not necessarily represent the views held by my employer, my partners or my clients. eDiscovery Today is made available solely for educational purposes to provide general information about general eDiscovery principles and not to provide specific legal advice applicable to any particular circumstance. eDiscovery Today should not be used as a substitute for competent legal advice from a lawyer you have retained and who has agreed to represent you.