Updates for May 2026:
- RelativityOne now integrates with Claude via Anthropic’s Compliance API, enabling streamlined, secure, and review-ready collections from Claude Cowork, Claude Chat, and Claude Code.
- With this launch, RelativityOne now offers native collection from all three leading enterprise AI tools: Claude, ChatGPT, and Gemini.
From emails and chat messages to collaborative documents, modern work has always produced a trail of discoverable data. Now, generative AI is creating a new category of business communication: conversations between people and machines.
Tools like Anthropic’s Claude, OpenAI’s ChatGPT, and Google’s Gemini are now doing far more than answering questions. They’re becoming part of everyday work across functions and industries. Product teams are using AI to draft code and accelerate development. HR teams create interview materials and offer letters. Marketing teams generate campaign briefs, customer communications, and press releases. And across the business, employees are turning to conversational AI to summarize meetings, answer policy questions, and help shape new ideas.
Each of those interactions leaves behind a record – one that may contain proprietary information, sensitive business context, or material that becomes relevant in litigation, investigations, or compliance matters.
Why Conversational AI Data Matters
A recent Gallup survey found that half of employed U.S. adults use AI in their jobs. As generative AI becomes embedded in everyday business workflows, the data it creates is becoming more varied, more sensitive, and more consequential.
Employees are using these tools for far more than brainstorming. They are:
- Asking strategic questions that involve confidential customer or financial information
- Drafting contracts, policies, proposals, and other business-critical documents
- Uploading files for summarization, editing, or transformation
- Generating code, content, and images that may raise important questions around ownership, authorship, and risk
AI interactions create digital records – including prompts, responses, uploaded files, and metadata – that can become key evidence in litigation, investigations, audits, and regulatory inquiries. Those records may be relevant to matters involving intellectual property, employment law, compliance, trade secrets, and breach of contract, and like any business communication, may need to be preserved, reviewed, and produced.
Until recently, however, many AI chat tools lacked the compliance features needed for defensible collection and review, leaving users with limited exports, inconsistent metadata, and no reliable path into e-discovery workflows.
We were eager to help address the gap. RelativityOne now offers native collection from all three leading enterprise AI tools: Claude, ChatGPT, and Gemini. Supported by our partnerships with Anthropic, OpenAI, and Google, legal and compliance teams can now access conversational AI data in a way that is defensible and ready for review.
This is an important step toward making conversational AI evidence as accessible, reviewable, and defensible as any other form of digital communication.
Claude Enterprise
Anthropic’s Claude has quickly become a major presence in enterprise AI, now accounting for more than a third of business AI usage across industries. As adoption grows, so does the volume of sensitive, business-relevant data generated within Claude. Strategic analysis, internal communications, and substantive work product are increasingly taking shape in this environment – making Claude data a growing source of potentially relevant evidence for legal, compliance, and security teams.
RelativityOne now enables teams to collect Claude Enterprise data directly from the source through Anthropic’s Compliance API. That includes full conversational content – prompts, responses, uploaded files, and project context – which can be brought into RelativityOne in a format built for downstream review. Once collected, the data is automatically converted into Relativity Short Message Format (RSMF), allowing reviewers to analyze Claude conversations with the same clarity and context they expect from any other message-based evidence.
The result is a more seamless path from collection to review: no manual exports, no lost context, and no formatting inconsistencies. Just structured, review-ready access to one of the fastest-growing AI data sources in the enterprise ecosystem.
Google Gemini
As generative AI becomes more deeply embedded in workplace productivity tools, Google Gemini has become part of the enterprise data landscape just as quickly. Integrated across Google Workspace, Gemini is increasingly used for document analysis, content generation, and decision support – creating a growing body of conversational data that may become relevant in legal and compliance matters.
With Collect in RelativityOne, teams can bring Gemini data into the same defensible workflows they already rely on for other enterprise sources. Prompts, responses, and associated metadata are collected in a clear, structured format and automatically normalized into RSMF, making Gemini conversations easy to review in context.
As AI tools continue to evolve and spread across enterprise workflows, legal teams need a consistent way to capture and review their output without reinventing processes every time a new platform gains traction. Support for Gemini is another step toward making conversational AI data manageable within established e-discovery workflows.
ChatGPT Enterprise
ChatGPT Enterprise is where this work began. As OpenAI became the first generative AI platform to show up at scale across enterprise workflows, it also became one of the first places legal teams encountered a new class of potentially discoverable data – one that traditional collection methods weren’t built to handle. Our July 2024 integration with Collect helped lay the foundation for bringing structure, defensibility, and review-readiness to this then-emerging data type.
With ChatGPT Enterprise collection in RelativityOne, teams can capture full prompt and response histories directly from the source, along with critical context such as user information, custom GPTs used, associated files, and even images generated through DALL·E. That richer context helps legal teams understand not just what was asked and answered, but the broader story around how AI was used in a business setting.
That depth makes the integration particularly valuable. Instead of relying on screenshots or partial exports, teams can collect ChatGPT Enterprise data in a format that supports meaningful review from the start – making it easier to assess relevance, preserve context, and move forward with confidence.
Building a Future-Ready e-Discovery Stack
Generative AI is reshaping how work gets done. Along the way, it’s creating a new class of business records that legal teams can’t afford to overlook. With conversational AI becoming so embedded in day-to-day workflows, the ability to collect, review, and, above all, understand that data in context is becoming a core part of modern e-discovery.
RelativityOne comes in to help by connecting directly to the AI platforms enterprises are already using and translating their output into formats built for review – without sacrificing context, defensibility, or scale.
Whether it’s responding to an employment matter, investigating a potential IP issue, or assessing facts in a product liability case, legal teams need a clearer view of the AI-generated content that could shape real outcomes. RelativityOne helps make that possible.
What’s Next
Conversational AI isn’t standing still – and neither are the workflows built around it. As more AI tools become part of everyday business operations, legal and compliance teams will need scalable ways to account for their output across custodians, matters, and data sources.
For organizations managing large volumes of conversational AI data – or any other electronically stored information – our new Collect API makes it easier to automate collection workflows end-to-end. Instead of manually setting up and running individual jobs, legal and technical teams can build consistent processes that scale, freeing up time for the work that requires human judgment.
Our focus remains the same: helping customers manage this next generation of data securely, defensibly, and in a way that fits the realities of modern legal work.
Get Started
Native support for Claude Enterprise, Google Gemini, and ChatGPT Enterprise is now live in Collect in RelativityOne. If your organization is already using these tools, now is the time to start incorporating conversational AI into your legal and compliance workflows.
Reach out to your customer success manager or check out our release documentation to get started.
