A beautiful, if unadvertised, annual tradition here in Chicago is stepping outside on a brisk September day, watching a water taxi float by, and inhaling a deep, cleansing breath of that invigorating autumn air.
Sure, sometimes that air smells a little like exhaust fumes or the Dunkin on the corner. But to us, it’s just a little fresher than usual.
At Relativity Fest this week, we’re celebrating a similarly energizing moment in the legal community—a moment that will transform our industry and move us all forward.
How did we get here?
To start, generative AI has come leaps and bounds over the last year. It’s gone from a prototype in a test environment to an entire category of products for helping e-discovery, legal ops, and many other teams work in new ways.
We’ve never needed that kind of accelerator more than we do right now. Huge, complex data stores and increased pressure to do more with less have left legal teams needing best-in-class and end-to-end tools to facilitate that work.
Fortunately, we’ve also seen an insightful new definition of our work enter the chat: Legal Data Intelligence. This framework unlocks new opportunities for each of us to go above and beyond e-discovery, while applying everything we’ve learned from the last 20 years in this discipline. As Relativity CEO Phil Saunders shared this morning, 38 percent of workspaces currently in Relativity are for non-litigation matters like regulatory requests and internal investigations. Clearly you all are doing bigger, broader work than you were a mere 10 or 20 years ago.
Within the Relativity community, your teams and ours are actively innovating to create a better experience and a more robust platform. We’ve made huge investments—more than $150 million worth this year alone —in our technology and our people to help catapult us all into that next chapter. We’ve also doubled down on our operations to make your experience smoother.
This work is paving the way for a new era of document review and an ever more capable pursuit of justice.
That was the theme of this morning’s Fest keynote. In case you missed it, here’s a recap of the biggest updates we shared.
Capitalizing on a Transformative Moment in Generative AI
As the face of review is changing, so is the nature of our work not just in e-discovery, but in the broader legal realm. One way our community is tackling this new frontier is with generative AI. After a whirlwind of interest and excitement over the last two years, we’ve now had a chance to put big ideas (AIdeas?) into action. The result is Relativity aiR. Have you experienced it yet?
If not, Fest is the perfect time to dive in. The truth is that generative AI has the power to revolutionize review for all types of use cases—and not just in how you conduct review, but in what you’re reviewing in the first place. Generative AI tools create human-to-AI agent conversational evidence; can respond to natural language text, audio, and video inputs; and will transform the way you search and analyze potential evidence.
Relativity aiR for Review is already the fastest and most accurate way to conduct document review in RelativityOne. And we’re just getting started: with aiR for Privilege launching later this year, and aiR for Case Strategy on track for early next year, you’re about to find more impactful ways to deliver better results for your clients and stakeholders than ever before.
aiR for Review and aiR for Privilege
The review solutions we’re introducing with aiR are poised to revolutionize review with powerful generative AI that delivers faster analysis, more accurate output, and defensible results. aiR for Review and aiR for Privilege will enable you to conduct document review in a fraction of the time while uncovering essential insights that drive faster decisions and more productive legal strategy.
As Elise Tropiano, senior director of product management and a true aiR extraordinaire, shared during the Fest keynote: we can’t wait for you to experience how these tools can transform your first-pass review experience. This traditionally messy, chaotic, slow process has long required the use of search terms, email threading, deduplication, and prioritization. Inevitably, there is a divergence of ideas that requires reconciliation via detailed QC reviews and knowledge sharing which must be done and dusted before you can move on to validation and pre-production reviews. Throw in rolling productions, and it’s a headache and a half.
But aiR can change all of that. With it, a single lawyer or subject matter expert can describe your matter, give that input to aiR for Review, and allow the system to do its work consistently and comprehensively in very little time (we’re talking minutes and hours, not days or weeks). The same expert can then review the results—complete with citations and potential counter-arguments—to validate aiR’s predictions.
And that’s it. That’s the workflow.
Setup is simple. Real-time metrics keep you informed of the project’s progress. You can start with a small subset of documents to fine-tune your prompt criteria, if you wish, and reuse prompt criteria from similar projects. And aiR for Review is available right now. Today. You can start using it immediately.
You’re probably thinking all of this sounds great—but you’re going to need to see some real results before you take the leap. Right?
aiR for Review just became generally available, but we have been collaborating with customers on its development for nearly 18 months.
As for in-the-field proof points? In one project with JND, aiR for Review found 90 percent of relevant documents in less than a day. aiR for Review has been utilized by dozens of customers, across hundreds of matters and millions of documents, in Australia, the UK, and the US.
Meanwhile, we’ve also been working on aiR for Privilege to help you protect critical information in your producible sets of documents. Privilege reviews are a different animal, where stakes are even higher and deep knowledge beyond the four corners of any given document is required to make the right calls.
This is why aiR for Privilege uses multiple AI technologies—machine learning, social network graphs, and generative AI—to offer a truly fit-for-purpose solution that can help you remove the noise in a data set, build context around each document, and make predictions about privilege automatically.
As with Review, aiR for Privilege backs up its predictions with a rationale for each—plus counterpoint considerations and a generated privilege log description that helps accelerate your production of that critical documentation, too.
aiR for Privilege is nearing release, with general availability targeted for next quarter. But it’s in a limited availability program right now; if you want to give it a try, reach out to your account manager to learn more.
aiR for Case Strategy
Next up in our generative AI toolkit is aiR for Case Strategy. Cristin Traylor, Relativity’s senior director of AI transformation and law strategy and a talented attorney with years of practice experience, got on stage to show off this new aiR solution.
On the one hand, case strategy is part of what makes legal work compelling: digging to the heart of a matter, uncovering the story to be told, and establishing how to zealously represent your client’s interests by building a strong narrative that speaks to the truth. But on the other hand, there are so many small steps—and so many roadblocks—that complicate this storybuilding exercise.
aiR for Case Strategy is designed to help flatten those roadblocks and enable faster, more meaningful intel on your projects—so you can focus on the higher-value work that those data insights inform. In aiR for Case Strategy’s Case Home interface, you’ll be able to create and explore fact lists, deposition outlines, witness summaries, and more. aiR will use relevant documents uncovered by review to identify key facts, people, timelines, and more—once again in minutes, not days!—and present a 360-degree view of all of this information to your case team.
The results are more informed attorneys who can reference a single source of truth to make strategic decisions about a case, quickly prepare for depositions, record and summarize transcripts, and so much more.
aiR on the Ground: Customer Stories
Let’s pause here for a quick refresher. At Relativity, our six AI Principles are:
- We build AI with purpose that delivers value for our customers.
- We empower our customers with clarity and control.
- We ensure fairness is front and center in our AI development.
- We champion privacy throughout the AI product development lifecycle.
- We place the security of our customers’ data at the heart of everything we do.
- We act with a high standard of accountability.
Part of delivering on every one of those principles is working with our customers to deliver the best possible products across all of aiR’s use cases and applications. Our Advance Access users’ feedback has been an essential part of our development process of each of these products—and their success stories are truly incredible.
Take a look at the video we shared during the Fest keynote to highlight those stories.
Streamlining the End-to-End Experience
Kyle Disterheft, a director of product management at Relativity and an all-around cool guy, joined us on stage in this year’s keynote to share an example of why we do what we do.
Thirteen years ago, as a member of our support team, Kyle took a call from a law firm customer. The user at this firm was calling Kyle from a beach, where he was troubleshooting a problem with a Relativity workspace for his team instead of spending his vacation time with his young son.
We know you do critical work, enabling the pursuit of justice alongside your teams. We’re here to support that work. To make it as straightforward and efficient as possible, so that you can find all the key insights you need—without wasting any of your most valuable time.
With this in mind, one of our top priorities this year has been making the end-to-end experience in RelativityOne simpler, faster, and better aligned with your goals.
Matter Management
Cross-matter reporting has a new home in RelativityOne’s Management Console, where you can report on workspace and case data over time—including project and processing statistics—without the need to run complex scripts. In 2025, you’ll find more data sources becoming available for this type of reporting—simplifying your billing workflows and allowing you to spend less time wrangling your data and more time using it to make strategic decisions.
Also coming soon, we’ve created a new AI-powered, in-app chat solution, AI Help, that enables you to ask questions and receive concise and accurate answers about how to run different workflows directly in RelativityOne. Use it if you’re new to the admin role, you took the RCA exam a few versions back, or you just need to find a quick answer without poring through documentation.
Data Uploading
Naturally, the first step in making document review work is making data visible in the first place—and that step hasn’t always been easy. Our team is always working to improve Collect and Processing in RelativityOne to make data uploading as painless as possible and give you a head start on your projects.
Today, Collect touches more than 150 terabytes of data each month—and usage is growing more than 370 percent year over year. And we’re adding new integrations all the time. For example, Collect now integrates with OpenAI’s ChatGPT Enterprise product, automatically collecting prompt interactions and converting them into Relativity Short Message Format (RSMF) for a near-native review experience.
This fall, we’re also launching updates to our Purview Sync application to help you more easily import data from Microsoft Purview Premium—including Copilot prompt data. And we’ve announced a partnership with Cellebrite, allowing users to initiate mobile collections from directly within RelativityOne and automate processing. Users taking advantage of this integration via our Advance Access program have seen an average collection time of just 70 minutes from initiation to processing.
And speaking of Processing, you’ll find lots of improvements there, too. We’ve doubled throughput for large jobs (with more improvements to come, amounting to over 5 terabytes per day by the end of the year); automated RSMF conversion for Cellebrite UFDR, Microsoft Teams, and Slack data; and are working on an “expert mode” for Import/Export that will give you push-button access to all those detailed processing settings in one place.
Review
Once your data is in RelativityOne, getting your hands around it—and gleaning insights from it—as quickly as possible is the name of the game. Review Center is now the home of all things document review in RelativityOne, and we think you’ll love the simplified review experience, real-time reporting, and always-on AI it offers.
For example, we released Coverage Mode earlier this year, which allows case teams to review the documents that help train AI fastest, first. This means it takes fewer human-coded documents for AI to identify the content that’s most likely to be relevant.
And soon, you’ll be able to save AI models to a new Relativity Model Library. This will allow you to reuse your models on similar matters—saving a ton of time and getting more value out of attorneys’ coding efforts.
Meanwhile, our new search infrastructure—currently in Advance Access—is massively scalable and optimized for interacting with message-level metadata. Starting next year, you’ll finally be able to run searches on deleted messages, active participants in a chat, and messages sent between specific people.
We’re also working on a generative-AI-powered, natural language search experience. You’ll be able to ask questions of your data set, like “What was Kyle Disterheft up to the night before the Fest keynote?” and derive meaningful insights quickly. You can also use it to run a complex search with everyday words instead of a bunch of technical syntax—so finding “all the text messages sent between midnight and 8:00 a.m. on May 5” is as easy as typing that phrase into RelativityOne. This is currently in the works, and we’re marching toward Advance Access early next year.
Expanding Beyond Litigation
Most of what we’ve discussed so far applies largely to litigation review, but we know that, now more than ever, a review for production isn’t the only kind of document review you do. The traction already gained by this year’s new Legal Data Intelligence framework is proof positive that legal teams’ use cases are expanding.
One area of particular popularity is contracts analysis, which many teams like yours already do today. We launched Relativity Contracts last year to help with these projects, transforming executed agreements into structured data. Next month, Relativity Contracts will leverage generative AI to automatically identify 16 of the most essential fields in a contract, including title, effective date, parties, and more—so you can easily compare details across contracts and better understand their context.
Another up-and-coming review use case comes in the aftermath of every organization’s nightmare: a cybersecurity incident. The unfortunate reality is that data breach victims were up 490 percent in just the first half of this year—making these tasks more common and imperative than ever.
We’ve launched Relativity Data Breach Response to enable you to conduct these critical projects within RelativityOne. Today, Data Breach Response expedites the delivery of entity reports that help you identify victims and enable faster, more informative notification workflows. The product leverages pre-trained machine learning models to find 120 out-of-the-box personally identifiable information (PII) types in a data set. A new impact dashboard gives you an at-a-glance view of impacted people, aliases, and the types of PII the system has found.
We’ve also introduced a new Entity Report for Data Breach Response, which is a first step toward simplifying the entity deduplication experience. Soon, you’ll also discover the ability to view and take action on conflicts—including confirming multiple email addresses belong to the same person.
The goal is to help you work smarter, not harder, on whatever types of projects you need RelativityOne to do.
CaiRry On Forward
Whether you grew up in the old or new school of discovery, this is an exciting time for our community. Our shared path forward will not be littered with wasteful paper or confusing bytes; instead, it will be paved with more of the fascinating insights that drew many of us to the intersection of technology and the law in the first place.
At Relativity, we get a warm fuzzy feeling whenever we look down that path and see the promising future ahead. But we can’t walk it, let alone help pave it, without you. Thank you for being an integral part of our team—and for making our community the treasure that it is. We can’t wait to show you what else we have in store.
Recommended for you
