Overall RelFest Impressions

The 25th RelFest Chicago was well attended, upbeat and polished. Relativity aiR ads greeted new (37%) and returning attendees as we deplaned in O’Hare. The Hyatt lobby and central venues were packed from morning til closing with networking opportunities. The event team dramatically expanded workshop and lab opportunities in response to last year’s demand. The partner exhibit booths seemed to have good traffic and I heard reports of much better engagement (prospects) from more educated customers.

Keynote Highlights and eDJ Takes:

Phil Saunders

Phil Saunders – CEO – Strategic Overview

  • Legitimate AI capability hype that has overshadowed the human impact.
  • Absorption Rate Reality – The news does not stop and can overwhelm practitioners. Choice overload.
  • $3.35B global legal tech investment this year.
  • 60x growth in enterprise data 2015-2030; IDC ‘s Gobal DataSphere.
  • Percentage aiR coding decisions grew from 4-57% in 18 months.
  • AI hallucinations- weak topic without a short term solution.
  • “Can I trust AI?” 35% share accuracy anxiety.
  • 25 years of Relativity innovation. Change is the new normal.
  • Humans are not built for constant change.
  • Innovate on a stable platform and UI to manage friction of change.
  • 27 advanced access programs.
  • 44% increase in RelFest workshops by attendee demand.

eDJ Take: Phil’s vision for Legal Data Intelligence continues to impress. The brand change from RelativityOne to Relativity aiR signals the market that AI is the new foundation of LDI. I would love to see the labs permanently online for Relativity customers to support adoption.

Chris Brown – CPO

Chris Brown – President – Building the Platform

  • Doubling down on ODA – Organize Discover Act.
  • Almost 100 PB customer data under management.
    • eDJ: While impressive, I wince at the thought of how much of that data is not relevant. The value lies in application of the tools on rich, curated collections.
  • Relativity aiR – AI platform for LDI.
  • Custom analyses- no code builder extension for custom classification and field extraction.
  • 2027- extend MCP agents to AI surfaces.
  • Agent harness model – data, capability, expertise/skills, knowledge supporting human decisions.

eDJ Take: Chris’s focus seems to be on AI enabled architecture and building a platform that meets users where they work. Recouping $200M in annual R&D requires practical infrastructure to support innovation from the entire Relativity community.

Lindsay Lanier – Sr. Director Product Management – Mature AI Workflows

  • Assist – counsel access to Early Insights reports to understand content in context.
    • AI analysis increased to 300,000 per report.
  • Expanded Collect sources.
  • Adding collection of AI conversations directly from AI services.
    • eDJ: AI prompts from investigation targets has been highly valuable in my recent support of client investigations.
  • Processing can reconstruct or reconcile versions in the context of communications.
  • MCP currently supports workflow and reporting actions, but not document access.
    • eDJ: Good. Better to get this right before AI agents can access sensitive customer data.

eDJ Take: As always, Lindsey’s keynote sessions focus on live navigation of Relativity’s cutting edge new features and UI enhancements. Sufficient to say that I want to get my hands on the new toys asap.

Chris Haley – Sr. VP, Product Management – AI Driven Review Enhancements

  • Review foundations
    • Classify – Relevance, Issues, Key Docs
    • Protect – Privilege, Confidentiality, Personal Data
  • Integrated review UI with aiR Review summary, topics, rationale, considerations and more all cross linked to support human decisions.
  • Expanding AI data types with customer analyses:
    • Image analysis and summary
    • Primary date extraction
    • Handwriting/Signature detection/extraction
    • Geographic location identification
    • Fake AI image detection
    • Certifications and education extraction
    • Medical record extraction
  • Review Center – workflow management across lifecycle that includes planning, analysis, extraction, validation and productions.
  • Matter Home- finally a repository for case docs, memos, notes, transcripts and more.

eDJ Take: Chris’s last preview of the Matter Home really caught my attention. I and many of you have been asking for a user area to share and organize workspace work product, review guides and more for a long time. It is great to see it on the horizon now that I am back to coordinating cases. Anything is better that trying to get counsel and PMs to use Teams.

Dorna Moini – VP Product Management – Introducing Gavel integration

  • Relativity acquired Gavel, AI-native automation of legal work product
  • Brings Relativity workspace content directly into Word to support drafting, edits, exhibit linking and more.

eDJ Take: This is the first extension of Relativity aiR workspace content to meet motion practice where it happens. It extends the reach and impact of discovery workflow output beyond the platform. No more copy/export/download. Another session that you had to see to grasp the potential impact. 

Elise Tropiano – Sr. Dir. Product Management – Introducing claiR

  • Relativity’s new claiR conversational UI allows legal teams to simply ask questions, explore processed collections and generate work product grounded in their workspace content.
    • eDJ: claiR finally tackles the ‘expert platform for experts’ adoption hurdle for non-technical counsel.
  • The AI responses are linked to relevant documents directly or with created searches. This gives practitioners the same interactive experience they are using with frontier AI services. It transfers those new skills directly into the Relativity UI.
  • claiR includes personal workspace and project folders that preserve context like ongoing Claude/Copilot/etc. projects. Natural chat conversations and saved skills.
  • Draft internal guides and versions with Gavel integration for editing.
  • Still dependent on the technical support teams to collect, process, analyze and otherwise enrich the underlying data matrix that answers are based in.

eDJ Take: claiR bring Assist into a much more familiar and functional interface. While Elise reassured PMs and other specialized support roles, I can see why many in the market may fear for their role’s ongoing value with claiR giving counsel dynamic data interactions directly without the request/fulfillment friction. I also wonder about the potential that Clair and similar tools may inadvertently create false confidence in counsel who lack broader technical understanding of collection composition, search limitations, metadata deltas and other challenges. AI can be smart, but it does not know what it does not know.  

RelFest Sessions – the few that I could make

13th Annual Judicial Panel

  • Still seeing hallucination in motions, but models are getting better at detecting them.
  • AI usage by litigators is happening and not commonly disclosed.
  • Study- 500 federal judge’s use of AI. Only 40% said they never used AI.
    • May be much higher usage among law clerks.
  • AI adds efficiency along with commensurate risk.
  • Seeing heavy AI usage from self-represented plaintiffs.
  • State court survey (Relativity sponsored) saw roughly 2x usage compared to Federal.
    • State judges have much less resources, so more likely causation.
    • Court staff policies seem all over the place (encouraged to banned).

eDJ Take: AI adoption in the courts is much faster than prior technology waves.

From First Signal to Final Decision: Navigating Internal Investigations in RelativityOne

  • Panel seemed to assume investigators know the players, have holds, collections and processed into RelativityOne.
  • Prioritization of comms, especially non-email.
  • Collapsing time to outcome driven by AI.
  • “Are there conversations that do not make sense?” The AI picks up across docs to catch code words.
  • Litigation has context and demands while investigations are usually much less defined.
  • Investigations have open questions that lend themselves to prompts.
  • Do not overcomplicate prompts. Keep them straight forward and goal driven.
  • Do not put conflicting issues or potential facts into the prompt. Let the AI come up with those.
  • Scenario models have to be revised when the issues shift after the seed sets.
  • Recommend hiding the AI reasoning from human reviewers to prevent bias and humans just following the AI lead.
  • “You need an expert in prompt creation.”
    • eDJ: This matches to recent studies and feedback. Prompt curation and training takes experience though the system is more approachable. Trust but verify.
  • “Do not leave you SME alone with the prompt.” They need technical support to measure effectiveness with sampling.
  • Case Strategy surfaces docs that counter the assertion and help preserve investigation integrity.

eDJ Take: The panel was informative but focused on very high risk matters or regulated clients with deep resources. I would have liked to hear more typical small HR, compliance investigations.

Peer Conversations

Peers are experiencing pressure to demonstrate ‘real’ AI driven overhead costs as opposed to potential discovery savings based on traditional review costs. Lots of stories of how AI cuts billable and response time in research, generating deliverables, drafting and such. Few peers want to cut their team or radically change traditional workflows to shift ‘ownership’.

Interesting stories of ‘vibe coded’ agentic discovery workflows using internal AI capabilities. The pride and excitement around these innovation projects reminded me of Cory Doctorow’s Makers. The token costs for these agents can be substantial but may deliver those overhead reductions management is demanding.

Law firms that invested in AI early on are struggling to decide whether to continue that investment in these boutique solutions targeting narrow research and operational user experiences in light of the evolving agentic capabilities of the Frontier models and the embedded AI capabilities in Relativity aiR and other platforms.

Collecting, processing full custodial collections is just not practical or efficient for the vast majority of usage scenarios. Even ‘unlimited AI’ licenses have volume and total token usage limits.

Like Microsoft in May, many companies have reigned back unlimited AI usage by knowledge workers in light of growing token bills. I have yet to hear of any firms passing general staff AI costs through to clients outside of approved discovery software costs. I have also witnessed client surprise at the Copilot Studio action-based costs on top of their flat $30/month/user Copilot fee. This was a hot topic amongst my more innovative peers deploying agents and MCP workflows.

There seems to be a demand for top eDiscovery experts based on the numbers of “Do you know anyone sharp I could hire?” conversations.

RelFest 2026 Press Briefing

I had a conflict this year and opted out of the Thursday press briefing. The esteemed Bob Ambrogi did his usual excellent write up Relativity CEO Says 2028 Server Deadline Stands; Will Prioritize Partnering with Frontier Labs Over Legal AI Companies.

Greg Buckles wants your feedback, questions or project inquiries at Greg@eDJGroupInc.com.  Reach out for a free 15 minute ‘Good Karma’ call if he has availability. He solves problems and creates eDiscovery solutions for enterprise and law firm clients.

Greg’s blog perspectives are personal opinions and should not be interpreted as a professional judgment or advice. Greg is no longer an investigative journalist and all perspectives are based on best public information. Blog content is neither approved nor reviewed by any providers prior to being published. Do you want to share your own perspective? Greg is looking for practical, professional informative perspectives free of marketing fluff, hidden agendas or personal/product bias. Outside blogs will clearly indicate the author, company and any relevant affiliations. 

Greg’s latest nature, art and diving photographs on Instagram.

 

0 0 votes
Article Rating