Last week\u2019s ILTACON, the annual conference of the International Legal Technology Association, brought what may have been the busiest news cycle I can remember for this conference, with a torrent of product announcements dominated by two themes: agentic AI and the Model Context Protocol. I have already covered the conference and several of the week\u2019s [\u2026]<\/p>\n
Last week\u2019s ILTACON, the annual conference of the International Legal Technology Association, brought what may have been the busiest news cycle I can remember for this conference, with a torrent of product announcements dominated by two themes: agentic AI and the Model Context Protocol.<\/p>\n
I have already covered the conference<\/a> and several of the week\u2019s biggest stories \u2013 including LexisNexis\u2019s Legal Intelligence Engine<\/a>, Thomson Reuters\u2019 launch of its Thomson LLM<\/a>, and Clio\u2019s hiring of Pablo Arredondo to lead a new judiciary initiative<\/a>. Now, as I dig out from a hectic week, I am rounding up here other notable announcements from the conference.<\/p>\n I\u2019ll do this in four parts:<\/p>\n Let\u2019s start with e-discovery.<\/p>\n Note: I used Claude to assist in creating the summaries.<\/em><\/p>\n E-discovery company DISCO<\/a> announced an enhancement to its gen AI-powered Auto Review product designed to remove the prompt-engineering hurdle from AI document review.<\/p>\n The new capability improves tag descriptions based on a user\u2019s plain-language feedback, runs tests on sample documents, and generates new prompts with improved performance metrics \u2014 letting legal teams refine their own AI review instructions without prompt-engineering expertise or reliance on outside legal service providers.<\/p>\n James Park, DISCO\u2019s director of AI consulting, called it an \u201ceasy button\u201d for complex reviews. \u201cWe\u2019re removing the technical trial-and-error of prompting to give litigators full access to the power and speed of AI review without relying on legal service providers.\u201d<\/p>\n The announcement followed DISCO\u2019s pre-conference launch of Advanced Research, an agentic AI application for fact investigation that uses multi-step reasoning to work through complex queries and generate memos, deposition outlines, and timelines.<\/p>\n \u201cAdvanced Research is a purpose-built eDiscovery tool for fact investigation that\u2019s designed to tackle complex queries using agentic \u2018deep thinking\u2019 reasoning to lean into challenging questions and deliver thorough responses to inform the best investigative strategy,\u201d the company said.<\/p>\n And those announcements follow one I covered early in August<\/a> of DISCO\u2019s launch of a unified litigation solution that combines the facts of a matter with relevant U.S. case law.<\/p>\n E-discovery company Everlaw<\/a> announced strategic partnerships and integrations with Thomson Reuters CoCounsel Legal, Harvey, Google Cloud\u2019s Gemini Enterprise for Legal, and Microsoft 365 Copilot, connecting those AI tools to what Everlaw calls its governed evidence foundation.<\/p>\n The integrations make Everlaw unique among e-discovery companies in allowing legal teams to access, search and analyze their evidence and data from partner AI platforms, the company says.<\/p>\n \u201cThe other leading platform\u2019s MCP offering is limited to project settings and admin tasks and does not provide access to evidence,\u201d a spokesperson said.<\/p>\n A Harvey user could search millions of documents in Everlaw to extract case-critical details, for example, while a Copilot user could draft deposition questions using facts organized in Everlaw.<\/p>\n Throughout, Everlaw remains the system of record for the underlying evidence, permissions and audit trail, the company says.<\/p>\n Founder and CEO AJ Shankar said litigation data is too complex for any \u201csingle silver bullet\u201d and that Everlaw\u2019s integration ecosystem gives customers freedom to connect their preferred AI tools to the evidentiary record.<\/p>\n The CoCounsel Legal and Harvey integrations are expected to be available to joint customers in fall 2026. The Gemini Enterprise MCP integration is in private preview, and the Microsoft Copilot connector is now live.<\/p>\n The announcement builds on Everlaw\u2019s previously announced work with Anthropic\u2019s Claude for Legal and its partnership with Legora<\/a>.<\/p>\n Nuix<\/a> announced a major gen AI expansion of Nuix Discover, its legal review and e-discovery platform, headlined by AI Chat, a conversational interface that lets reviewers, attorneys and investigators interrogate case data in natural language and receive synthesized answers with citations to documents and a complete audit trail.<\/p>\n The broader release also includes Document Summaries generated at ingestion, Semantic Search for locating material by meaning rather than exact terms, Similar Documents for surfacing documents that share characteristics, and AI Clustering and Visualizations that map document collections to give review teams an early read on key issues and outliers.<\/p>\n Nuix says the capabilities are built into both early case assessment and review workflows.<\/p>\n In the announcement, Ilona Meyer, executive vice president for Nuix Discover, said that answers are grounded in case data and auditable \u2013 the standard, she said, that the legal profession demands.<\/p>\n AI Chat is available now through an early adopter program, with general availability in December 2026. The other gen AI capabilities become generally available on Sept. 4.<\/p>\n Relativity<\/a> announced that RelativityOne now integrates with Google Cloud\u2019s Gemini Enterprise for Legal through the Model Context Protocol, as part of Google\u2019s launch of the new legal-specific offering.<\/p>\n Gemini Enterprise for Legal is one of the first in a series of packaged industry solutions built on the Gemini Enterprise platform, delivering domain-specific AI capabilities that may include tailored agents, specialized skills, data connectors, and model optimizations for legal practice.<\/p>\n The Relativity MCP now connects to Gemini Enterprise, adding a conversational agentic execution layer on top of RelativityOne, the company said. This lets system administrators \u2013 including litigation support teams, legal operations, service providers and developers \u2013 stand up new matters, align workspaces to the structure of their data, and manage access and workflows through natural language.<\/p>\n \u201cOur goal in connecting RelativityOne directly to Gemini Enterprise for Legal is to make leveraging advanced AI both practical and actionable for legal teams,\u201d said Chris Brown, Relativity\u2019s president. \u201cThis integration empowers users to orchestrate matters and workflows inside RelativityOne, their sensitive data never leaving the platform and the substantive analysis staying within Relativity aiR.\u201d<\/p>\n\n
DISCO Enhances AI-Powered Auto Review and Launches Advanced Research<\/strong><\/h2>\n
Everlaw Announces Four AI Partnerships, Positioning Itself as the Evidence Layer<\/strong><\/h2>\n
Nuix Expands Generative AI in Nuix Discover, Debuts AI Chat<\/strong><\/h2>\n
Relativity Integrates RelativityOne with Google\u2019s Gemini Enterprise for Legal<\/strong><\/h2>\n