AI has overtaken crypto, NFTs, the metaverse, and supply chain snarls as the subject of practically every business-related news article, blog post, or podcast episode. AI has been around for a while (you’ve been using spell check in your word processing software since your first job, right?), but ever since OpenAI released ChatGPT in 2022, you can’t escape discussions about how AI is going to “change the game.”
One of those “games” is the legal industry. Sure, AI is going to make briefs and contracts easier to research and write, but we believe AI in eDiscovery will be the most impactful application of AI in the legal industry.
AI will help legal teams manage and review an ever-increasing amount of data even more accurately and quickly than they can right now using today’s eDiscovery tools. With organizations always wanting to reduce the amount of money they spend on litigation, AI for legal discovery will bring new levels of automation, efficiency, and accuracy to eDiscovery.
A longtime thorn in the side of document reviewers and their colleagues is identifying privileged or confidential information in ESI for redaction or withholding of the document. For years, review teams needed to read documents line-by-line to identify privileged materials. However, an AI legal assistant can introduce automation to the review process.
Logikcull’s new Suggested Tags feature uses AI to suggest tags for documents, such as “privilege,” and identify documents under review that might fall within that tag.
From Clueless to Clued-In: The Magic of Suggested Tags
Suggested Tags uses rule-based and machine-learning tools to suggest privilege tags for documents based on characteristics like common terms used in documents to signify that they may be privileged, such as “Privileged & Confidential,” or “Subject to Attorney-Client Privilege,” associations with documents already tagged as privileged, and patterns in tagged documents.
Each suggestion comes with a confidence score reflecting how sure the platform is that the document is privileged or contains privileged information. Besides the confidence score, the Suggested Tags feature also “shows its work” like you had to do in math class. When Logikcull suggests a tag, it provides a tooltip explaining why it made that suggestion.
Did we mention the Suggestion Tags recommendation engine also gets better as review teams tag more documents as privileged? Suggested Tags learns which documents in a project are likely to be privileged and continuously refines its suggestions and confidence scores.
With Suggested Tags, review teams will benefit from better data organization and will speed up their document retrieval efforts by knowing at a glance which documents may be privileged. Once they toggle the privilege tag, they can retrieve all potentially privileged documents, allowing reviewers to immediately access documents that need to be checked for confidentiality or privilege.
Embrace AI in eDiscovery with Logikcull
Suggested Tags shows the benefits of putting AI and machine learning to work in eDiscovery—and previews the sophisticated workflows AI can tackle in eDiscovery and throughout the litigation process:
- The tool looks for words it knows are used often in privileged documents and communications to determine if a particular document is privileged
- Based on what it learns about the characteristics of documents tagged as “privileged,” it searches for other documents in the set that may be related to privileged documents and links them to each other so reviewers can easily find related documents
- It judges how certain it is that a tagged document is privileged, gives the user a score regarding its level of certainty, and explains in plain English why it tagged the document
- Finally, the tool continuously refines its understanding of privileged documents within a particular document set, including each time a reviewer confirms a tag or manually tags or untags a document
Suggested Tags is one of the first AI-powered tools in the Logikcull platform, but it won’t be the last. We see the future and how, in that future, AI and machine learning will only make the eDiscovery process quicker, less costly, and more accurate for legal teams.
We’re working on introducing other AI-powered tools, including Logikbot AI, a document review tool that integrates ChatGPT to identify relevant documents or suggest redactions with natural language prompts, and Automations, which will enable users to configure Logikcull to automatically perform routine tasks like culling, tagging, and redacting as documents are uploaded into a project.
AI in eDiscovery will be a powerful tool to help legal teams improve efficiency and accuracy while reducing the time and money they spend on the discovery process.
Over the next few months, Logikcull will introduce AI-powered features into the platform, beginning with the new Suggested Tags tool. By automating the process of identifying potentially privileged documents, Suggested Tags helps review teams quickly get to the documents that need their attention the most. Confidence scores, explanations for why documents were tagged as privileged, and an always-improving recommendations engine show how helpful an AI legal assistant can be in the discovery process.
With apologies to our good friends working hard in other corners of the legal tech world, AI for legal discovery is the AI/legal mashup that is going to make the biggest impact on the everyday lives of in-house and outside lawyers and their colleagues. AI in eDiscovery will help legal teams review documents more effectively than before and in a fraction of the time it once took them.
With Suggested Tags and our forthcoming AI tools, Logikcull is leading the way with AI in eDiscovery. Many legal teams want to integrate AI in the discovery process. If they’re Logikcull users, they’ll already have access to our groundbreaking AI.
If you're interested in incorporating AI in your eDiscovery process, Logikcull can help. Schedule a demo today to see how you can put Suggested Tags and our other AI-powered features to work.