Already using Harvey? Here’s how Claira can extend AI into your document review
By Kate Clark
In part one of this article, we explore how Claira, working within Nuix Discover, can build on the investment firms have already made in Harvey.
:quality(80)/f/285684285272562/1920x1080/6c2589faf7/1.png)
Law firms are investing in AI tools such as Harvey to help their lawyers work more efficiently, analyse information and prepare legal work.
But what happens when a matter moves from understanding the issues to reviewing thousands of documents, making relevance and privilege decisions, and preparing documents for production?
This is where Claira, working within Nuix Discover, can build on the investment firms have already made in Harvey.
Rather than choosing between the two, I think the more useful question is how they work together across the litigation process. The starting point is understanding what each tool does well, where the work takes place and how the outputs can be used in the next stage of the matter.
Different strengths, complementary roles
Harvey and Claira both use AI to help legal teams analyse documents, but they are built for different parts of the litigation process.
Harvey supports legal analysis and case preparation. It runs on its own platform. Lawyers can use Harvey to research legal issues, analyse document collections, extract information, develop chronologies, prepare witness materials and assist with drafting. Its document review capabilities also allow teams to analyse and organise information across large document sets.
Claira brings AI-assisted review directly into Nuix Discover. It works as a panel inside Nuix Discover so there is no data to migrate. Teams apply review criteria across document collections, assess relevance and privilege, and record coding decisions and supporting reasoning within the review platform.
Defensibility differs across the platforms. In Harvey, you will need to manually track and document prompt development. Users must maintain a record of the prompts used to generate outputs, changes made over time and the reasons for those changes, so that this information can be provided to the Court if required. This creates an additional administrative burden and a risk of incomplete records if prompt changes are not consistently document. By comparison, Claira tracks prompt development throughout the process and can produce a report for the Court if required.
One of Claira’s capabilities is Bulk Scan, which allows teams to test and refine review instructions against individual documents without token usage before applying them across larger document populations.
Importantly, Claira doesn’t simply classify documents. It records the reasoning behind each review decision, giving reviewers information they can assess, verify and use as part of their quality control process.
Where the differences show up
The distinction becomes particularly important when a matter moves towards disclosure or production:
Privilege: Claira makes first-pass privilege calls and records the privilege type and basis in Discover fields, where the privilege log is built. Privilege flags raised in Harvey are useful for analysis, but they still need to be carried into the review platform before production.
Audio, video and images: Claira scans these file types directly, in bulk, inside Nuix Discover, alongside the rest of the review population.
Decisions stay with the documents: Claira’s coding and reasoning sit against each document in Nuix Discover, connected to the coding fields, document families and quality control processes used to prepare materials for production.
The opportunity isn’t to replace Harvey with Claira. It’s to use Claira to extend the benefits of AI into the document review and production workflows that litigation teams already rely on.
What about document volume?
Document volume is certainly relevant, but I would avoid drawing a simple line between using Harvey for smaller document sets and Claira for larger ones. Harvey can analyse large document collections, and both tools offer capabilities that can assist with document reviews.
For me, the more useful distinction is what the team needs to achieve, where the review decisions need to live and how those decisions will be carried through to production.
Where a legal team already uses Nuix Discover to manage document families, review fields, quality control and production, Claira provides an opportunity to bring AI-assisted review into that existing environment.
That can be particularly valuable when review criteria need to be applied consistently across a document population, and the results need to remain connected to the production workflow.
No extra platform commitment
Adding Claira doesn’t require another firm-wide platform commitment. It is priced per case per month, and can be turned on and off when needed, so it can be scoped to the matters where structured review is needed.
Token usage
With Claira, token usage is transparent.
Before starting a bulk scan, you can see how many tokens the job will use and your remaining balance. You can also test and refine your prompts in the free testing area before committing tokens to a larger review. That gives legal teams greater control over usage, costs and how they approach the work.
Building on what’s already working
Harvey can help lawyers develop their understanding of a matter. Claira applies that understanding across the documents being reviewed in Nuix Discover. The evidence identified through that process can then inform further legal analysis and preparation. For firms already investing in Harvey, Claira offers another way to get value from AI across the litigation process, without replacing the tools or workflows that are already working well.
In part 2 of this series, I look at how to set up the workflow between the two tools, from the first handover to final production.
At eDiscovery Collab, the exclusive provider of Claira in Australia and New Zealand, we are helping legal teams explore how Claira fits alongside the technology they already use.
If your firm is using Harvey, and you’d like to understand how Claira could enhance your existing document review workflows, I’d be happy to show you.
:quality(80)/f/285684285272562/756x667/c4c045a5f7/kate-clark.jpg)
Kate Clark
CEOKate is a senior eDiscovery expert with over 30 years of experience dedicated to simplifying complex legal data and streamlining operations for lawyers.
View full profile:quality(80)/f/285684285272562/1920x1080/ffa6ca44cb/2.png)
:quality(80)/f/285684285272562/4244x2345/c94015f49b/istock-1474113759-2.jpg)
:quality(80)/f/285684285272562/908x641/d93101de75/claira-blog.png)