Is agentic AI document review transformative for use in complex litigation? One company says “yes” and backs it up with TONS of real-world examples.

Just before Legalweek, Syllo released a new white paper titled Agentic AI Document Review Is Transformative for Complex Litigation (available here and also in PDF form here). The paper argues that traditional approaches to document review in complex litigation, including outsourced managed review and non-generative Technology-Assisted Review (TAR), have significant limitations in terms of cost, accuracy, granularity, and adaptability.

Is generative AI the answer? According to Syllo, while GenAI offers promise, linear applications of LLMs to large datasets face challenges such as prompt overload, limited issue codes, and lack of cost-effective adaptability.

So, what is the answer? Syllo presents its agentic AI document review as a novel solution that overcomes the limitations of traditional approaches to document review (like outsourced managed review and TAR), as well as linear applications of LLMs. Syllo’s system coordinates multiple LLMs to autonomously manage and execute document review tasks based on user-defined guidelines. As they discuss, their methodology offers several key benefits, including:

• Dynamic Resource Allocation: Syllo intelligently allocates more powerful LLMs to complex documents and simpler LLMs to straightforward tasks, optimizing cost efficiency and accuracy.
• Unlimited Issue Coding: The agentic approach can cost-effectively handle an unlimited number of issue codes without causing prompt overload or impacting accuracy. The system also provides explanations for each applied label.
• Swift and Cost-Effective Adaptability: New legal issues can be integrated seamlessly without retraining or reprocessing the entire dataset. Prior analyses can be leveraged for subsequent targeted queries, and coding refinements are significantly cheaper.
• High Accuracy: Empirical validation in live litigations demonstrates consistently high Recall (average 97.8%, median 99.4%) and Precision (median 85.9%, average 79.7%) in responsiveness reviews.

But “where is the proof”, you say? Without a doubt, the most compelling part of the paper is the “Empirical Validation” section which not only discusses how “Syllo’s AI systems have been used on active matters since 2023” but also provides several real-world examples of its use by real litigators, with actual quotes from them.

Syllo provides ten real world examples in all: four examples of its use in responsiveness reviews, three examples of its use in finding hot documents and three examples of additional applications in litigation. Each of the examples has a quote from an actual litigator on their use of Syllo and the benefits achieved. The white paper lists the names of 25 litigators in all as “practitioner contributors” to the white paper from seven different firms.

Here’s a quote from one of the litigators: “Syllo’s automated document review is reliable and provides unrivaled transparency into specific document characterization,” said David Stanton, a partner at Pillsbury.  “Far from being a ‘black box,’ the tagging rationales applied by Syllo let us see why particular tags were applied to specific documents.  This enabled workflows to adjust, optimize, and confirm the instructions we provided, and allowed us to very quickly leverage the insights we gained from its use.”

Syllo’s conclusion to the paper? “When guided by sophisticated litigators, agentic document review provides a powerful strategic advantage in complex investigations and litigations.”

Is the white paper promotional in terms of promoting Syllo’s agentic AI document review approach and solution? Sure. Are there counterarguments that can be made to at least some of their positions? Perhaps. But the empirical validation provided by them in this white paper in terms of 25 litigators across seven different law firms illustrates plenty of real-world results using their solution. That’s compelling.

Last week at Legalweek, I had several discussions with people about the barriers to GenAI adoption. Trust in the technology was a top barrier, especially given all the coverage of lawyers submitting filings with fake case citations (which is an “apples to oranges” comparison to use cases like the ones mentioned by Syllo in their white paper).

Real-world examples of successes with the technology are needed to overcome all the “bad news” stories we’ve been hearing (even those “bad news” stories they’re not really applicable to the use cases discussed above). Syllo’s white paper on their approach to agentic AI document review gives us plenty of real-world success examples. Check out their white paper here or in PDF form here – you don’t even have to provide your information in a landing page to check it out!

So, what do you think? Do all the real-world examples discussed in this white paper impact your thinking regarding agentic AI document review? Please share any comments you might have or if you’d like to know more about a particular topic.

Image created using GPT-4’s Image Creator Powered by DALL-E, using the term “robots reviewing documents on computers”.

Disclosure: Syllo is an Educational Partner and sponsor of eDiscovery Today.

Disclaimer: The views represented herein are exclusively the views of the authors and speakers themselves, 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.