Your single source for new lessons on legal technology, e-discovery, and the people innovating behind the scenes.

How (and Why) to Apply AI to Your Next Prior Art Patent Search Project

Drew Deitch

Patent attorneys, picture this: it’s 6:01 p.m. on a Friday, and a new matter hits your desk. It could be a pitch for a new client, or potential litigation threatening an existing client.

Either way, your first step is likely to be a prior art search. But—as is often the case—you don’t have the time or budget to outsource a prior art search project right now. And you need some strong art by Monday to show your client. Ready to spend your weekend cruising Google Patents or (gulp) the Patent Office’s online search tools?

There’s a better way. Bring the power of artificial intelligence to your matter. Early-stage search is the perfect use case for AI to accelerate your work, and the perfect opportunity to start making technology platform decisions and building a data strategy that can benefit your matter from beginning to end.

So, if you have an urgent prior art search project and you want to use technology to help, what are your options? Before we dive into the details, let’s revisit a few foundational concepts related to patent and intellectual property work:

  • Not a litigator? Not a problem. Patent prosecution attorneys and other strategic IP counsel can just as easily run a quick, affordable search in Relativity Patents. The same benefits apply, especially for time-sensitive, cost-sensitive, or early-stage search projects.
  • Where does e-discovery come in? Prior art search generally begins before discovery in patent litigation. The earlier you can get a jump on your search—and the more closely you can tie your search strategy to your overall data and discovery strategy later in the matter—the more efficient you’ll be.
  • Lit support teams, we’re talking to you too. Patent attorneys may or may not be a group you work with today. But if not, they should be! They have technology challenges just like any other group. This may be a new attorney population you can support, a new workflow you can assist with, and—if you’re a cost-recovery or revenue-generating shop—another high-value task to bring in-house. And as we mentioned above, pre-litigation prior art search is a perfect way to create the matter in Relativity much earlier, giving your team more notice on what the attorneys are working on—and what data may be coming into RelativityOne down the road.

Now that we’ve established some background, it’s time to talk about how AI can be the thread that ties all these principles together, especially when paired with smart workflow and powerful review tools. As we’ve discussed many times on this blog, the more attorneys can use AI to amplify their work, the more they can focus on high-value, strategic tasks, while automating away the tedious and the error-prone.

Many attorney workflows benefit from AI in this way, and thanks to the ever-growing volume of patent data, prior art search is no exception. AI-powered prior art search from Relativity Patents can accelerate your workflow, help you find better and more extensive results than you would on your own, and just maybe … save your weekend!

Where to Begin with AI for Patent Search

Okay, back to our example scenario. It’s Friday at 6:05 p.m. The initial panic has subsided. Now you need to decide what to do next. Will you wade into the prior art search yourself? Try to find an outsourced search vendor to do a rush job (this can still take up to a week)? Will you do both in parallel?

Just as a strong early case assessment strategy is essential during e-discovery, a strong early-stage search strategy is essential for patent search or defensive patent litigation projects. But they’re cumbersome. Early-stage prior art search projects for clients are often performed by attorneys, so budget is tighter than outsourced projects. And pitches are likely not billable at all.

The good news? Pre-litigation searches don’t always need to be exhaustive. Most early-stage searches require only a credible, instructive set of references to establish a baseline. Which elements of your search project will be easier to find, and which will be trickier? The faster you establish this baseline, the more you can shape and guide the more intensive prior art search that may come later in the project (whether you’ll do the search yourself or outsource to a firm).

Because both the budget allocated and the bar for inclusivity are lower, these are the perfect projects for skeptical teams to pressure test AI-powered search platforms.

At many firms, attorneys turn to Google Patents as the go-to tool when performing their own searches. And Google Patents is great! But it’s great for looking at a small handful of patents at time. When you need to review more than a few references, Google becomes unwieldy. Users can get lost in a sea of tabs and forget which references they've already read. And there is no easy way to save good references or highlight and comment on what you’ve found.

This is where a more powerful solution shines. Relativity Patents is specifically designed for both prior art search and full-fledged patent review—with machine learning integrated throughout the workflow to lighten the load. Users enter a natural language prompt and instantly see a ranked list of high-quality prior art.

From there, Relativity Patents makes it easier to review each reference by jumping directly to the most relevant paragraphs in each patent. Users can review more quickly by focusing only on the most relevant sections. Attorneys can then export findings to share with the client or a search firm, or can begin the next phase of their project by starting to build work product (look out for a future post for more on claim chart automation).

Jump Right in with Relativity Patents

There you have it! With the help of machine learning, Relativity Patents saves time and money on quick searches, while improving quality—potentially saving your weekend as well. And with our new and improved global search, which provides instant results across all worldwide patents, you can run your project faster than ever.

If you’ve been wondering how AI can aid your IP practice, look no further. Prior art search is a great place to start. Clients also love to hear that their counsel is using cutting-edge tech to deliver better outcomes, affordably. So, what are you waiting for? Give Relativity Patents a try by requesting a free trial


Artwork for this article was created by Kael Rose.

Drew Deitch is commercial lead of Relativity Patents, where he helps patent attorneys and searchers work faster with AI. He's been at Relativity for 10 years, in roles supporting our strategic partnerships, executive team, and product strategy.

The latest insights, trends, and spotlights — directly to your inbox.

The Relativity Blog covers the latest in legal tech and compliance, professional development topics, and spotlights on the many bright minds in our space. Subscribe today to learn something new, stay ahead of emerging tech, and up-level your career.

Interested in being one of our authors? Learn more about how to contribute to The Relativity Blog.