GenAI Prompting for In-House Attorneys

Lawyers hate surprises. In-house counsel responsible for litigation strategy want to know what’s in their document set – good or bad – well before it gets to the other side.  

Document review, whether for a government records request or a litigation-related production, must be thorough, fast, and accurate. But the volume and complexity of data in today’s digital business environment make effective review harder than ever, and expensive. Many in-house law departments have to pay a lot of money to outside counsel to do the heavy lifting. 

This is where generative AI in discovery is already much more than hype. Responsibly designed GenAI features that rely on the use of LLM prompting – such as Everlaw^AI Assistant’s Coding Suggestions – can augment the work of humans by speeding up review determinations, surfacing hard-to-find insights from vast document sets, and serving as quality control.

Finding and Prioritizing Documents Faster and More Accurately

Counsel in charge of litigation strategy and those managing law firm relationships want to stay ahead of the information curve, not be forced into a reactive mode. 

Using prompts and coding suggestions can expedite fact finding to help the team understand the contours of a matter – including what issues are at play, how to characterize those issues, and the types of documents that might address them.  

AI-powered coding suggestions also serve as a backstop in quality control. Acting as a second pair of eyes, coding suggestions can flag inconsistencies and items that human reviewers may have missed, boosting the effectiveness of both in-house lawyers and their outside counsel.

“You want to know before the other side shows up in a deposition or a hearing what’s in your documents,” says Chris Rauen, a senior solutions architect at Everlaw and former practicing attorney. “This tool helps you discover that information more quickly and efficiently and manage that risk intelligently.”

Getting the Most Out of Coding Suggestions with Everlaw^AI Assistant

Coding Suggestions uses GenAI to evaluate documents against user-defined coding categories. Everlaw^AI Assistant proposes appropriate codes for each document, providing rationale for its suggestions and citing to the section of the documents that support each suggestion. Legal teams can then apply, remove, or replace codes directly in the review assistant. 

The AI streamlines the coding process by offering intelligent, explainable suggestions for document classification across various criteria.

To get the most value out of Coding Suggestions, attorneys need to know how to prompt the AI. 

With a talent for using language with precision, a skilled lawyer is in the perfect position to get good results. That’s because generative AI systems deliver the best results when it receives a specific and detailed explanation for the context and information being sought. Think of it as an intern or associate who needs to understand the topic to perform the task.

Read more about how Everlaw^AI Assistant’s Coding Suggestions feature works and tips for writing strong prompts in The Lawyer’s Guide to Prompt Whispering.

Managing Risk in a Range of Litigation and Investigations

GenAI-powered coding suggestions capability can accelerate time-consuming and tedious processes across a variety of matters.

HR and Employment

GenAI coding suggestions can quickly prioritize and flag documents containing potential discriminatory statements, evidence of retaliation, harassment, and other communications related to workplace complaints or discrimination allegations.

Antitrust

The technology can swiftly identify and highlight documents discussing pricing strategies, market share data, competitor communications, and potential anti-competitive practices.

Intellectual Property

GenAI coding suggestions can efficiently flag documents indicating ownership issues, prior art, copying indicators, and other evidence of potential infringement or problematic use of intellectual property.

Commercial Litigation

The system can rapidly prioritize contracts, financial records, correspondence between parties, and documents detailing business transactions or breaches of agreements.

Environmental Litigation

GenAI coding suggestions can expedite the identification of documents related to environmental impact assessments, compliance reports, hazardous material handling procedures, and communications about potential violations or incidents.

“This tool helps you discover the information more quickly and efficiently and manage the risk intelligently.”

• Share this quote:

With its ability to swiftly analyze documents and propose relevant codes, Coding Suggestions capability accelerates the review process, letting lawyers focus on higher-value work.

Putting LLM Prompting to Work for In-House Legal 

These days, human review can only get legal professionals so far. A lawyer may have seen a document and produced it, but the potential seriousness and impact of the information can still go unnoticed. 

Everlaw^AI Assistant’s Coding Suggestions is a powerful AI tool that gives teams an advantage. These tools can prioritize documents for review with greater speed and accuracy and flag those that may be particularly important. They build in an extra quality control element, serving as a second pair of eyes on issues that may have been overlooked. 

By crafting thoughtful and nuanced prompts, attorneys can guide large language models to generate the most valuable work product. No legal team will ever read the billion emails an organization generates, but backed with the latest AI tools, they increase the chance of identifying the potentially problematic ones and manage litigation and other business risks more effectively. 

To learn more about Coding Suggestions and Everlaw^AI Assistant, request a demo today.

Petra is a writer and editor focused on the ways that technology makes the work of legal professionals better and more productive. Before Everlaw, Petra covered the business of law as a reporter for ALM and worked for two Am Law 100 firms.