Ropes & Gray, an early adopter of AI in the legal industry, is developing a proprietary transactional diligence tool with OpenAI. The goal is to help lawyers maintain the same quality and consistency of advice while moving at a faster pace, especially for data-heavy work under tight deadlines. Josh Lichtenstein, partner and head of the ERISA fiduciary practice, says AI is one of the key ways—if not the only way—to do that.

Transactional diligence is the buyer’s pre-acquisition investigation into the seller’s financial and legal claims. It usually requires an entire team of lawyers to parse enormous numbers of documents, analyze patterns, and put together a bid in days of intensive work. Ropes & Gray’s tool takes a first pass by combing through thousands of pages, identifying relevant documents, and automatically grouping them by themes and issues. It inventories and categorizes the data room so lawyers can review full relationships rather than isolated documents, and it produces summaries plus a completed report in Ropes & Gray’s signature format. Lawyers can inspect the grouping and categorization to confirm nothing is missed.

The initial development focused on material commercial contracts. In working sessions, lawyers compared test outputs with human-authored diligence reports from completed deals, then refined the rubric and report format based on what drafts got right and where they fell short. The tool uses a consistent query and framework that mimics the firm’s own workflow: initial reviews, summaries, then higher-level reviews. Ropes & Gray describes the output as comparable to a capable junior-to-mid-level associate, giving lawyers a high-quality first pass.

Lawyers still review everything before delivery. As Lichtenstein explains, anything delivered to a client must be something the firm can stand behind 100%, so the tool is designed to let lawyers easily review interim summaries and actual documents. With the first pass handled, lawyers spend more time reviewing and refining, which leads to a more carefully crafted final judgment. Ropes & Gray has publicly described the ambition of producing a detailed, issue-level diligence report in hours rather than weeks, and says that with tools like this one and GPT-6 Astra, client problems are being solved far faster than before.

The firm also draws leadership lessons from the work. Lichtenstein recommends experimenting with AI across many parts of life, noting that mistakes are where he learned to turn AI from a toy into a powerful professional tool. Howard Glazer, partner and head of practice AI, stresses putting clients first: the opportunity matters only if clients are the ones who win. Looking ahead, Glazer believes the next stage of AI will create collective knowledge inside a firm of 340 partners and 1,500 lawyers, giving people access to knowledge from client meetings and past projects. That, he says, is not just saving time or money; it is using AI tools to directly create value.