In the AI Startup of the Week, the editorial staff of ai.nl is featuring promising AI startups, their innovations, solutions and challenges. In this twenty-fourth episode, we are taking a look at Amsterdam-based legal tech startup Uncover. The startup founded by former litigators wants to help lawyers win cases using AI.
One of the earliest promises of artificial intelligence is its ability to help with industries where there is paperwork involved. AI can be used as a combination of technologies that readily transforms these documents into actionable insights. One industry where documents pile up rather fast is in the world of law. Uncover, based in Amsterdam, has found an opportunity to not only fix this influx of documents but also transform it.
There are a number of things that AI can do well and one thing that it does absolutely well is classification. For lawyers, having all of their case files classified based on category or subject matter can make a major difference. As former litigation lawyers, Caroline Korteweg and Ingrid van de Pol-Mensing understand the impact of AI and are uncovering it for the world with their startup.
Personal experience looking for documents
The origin story of Uncover is similar to that of most startup founders – it is rooted in personal experience. As litigation lawyers, Caroline Korteweg and Ingrid van de Pol-Mensing have faced the challenge of spending late night hours physically looking for documents. These documents also guide them to rearrange the exhibits and craft event timelines accurately.
So, it is not just the significance of these documents that matters but also their easy access. With litigation lawyers having already figured out the significance of case documents, the co-founders of Uncover are trying to help them with easily accessing the documents. On its website, Uncover makes a bold claim of a world where these litigation lawyers become “one with the case file.”
The final goal is to create a solution that helps litigation lawyers to focus on winning the case without worrying about the documents. To achieve this goal, Uncover is being offered as both a mobile as well as desktop application.
Finding structured data from unstructured information
Uncover offers a solution that utilises uniquely built algorithms to help attorneys structure all the unstructured data from their case file. They can then work with these materials to prepare for their case and even use the same as a reference material throughout the litigation process. The data can also be used while collaborating with colleagues or clients on the case. At the heart of this technology is one of the most popular AI techniques called Natural Language Processing (NLP).
Caroline Korteweg and Ingrid van de Pol-Mensing have combined their decades of experience as litigation lawyers and turned them into an AI engine. In addition to NLP, they also deploy deep learning techniques to reach conclusions and deliver optimised results.
For building their AI engine, Uncover is partnering with Slimmer AI, a Groningen, the Netherlands-based AI B2B venture studio that has helped the likes of Biller AI to build a revolutionary AI product.
“Today, lawyers spend roughly 30-40 per cent of their time ploughing their way through documents focusing on the factual information, trying to make sense of the case file. It is fundamental but inefficient and time-consuming work, resulting in lawyers longing for more hours in a day while clients are demanding more efficiency. We searched long and hard for solutions in the market to attend to these issues but came up empty-handed: none seem to have been built with a litigators’ approach in mind. Uncover will be exactly that: a tailored solution by litigators for litigators,” says co-founder Caroline Korteweg.
How does the Uncover platform work?
The platform built by Uncover for litigators works on five guiding principles: Structure, Timeline, Draft, Search, and Collaborate.
- Structure: This is the most critical element of the Uncover platform and also the fundamental block. The platform takes case file documents and structures them using fast and automatic classification. This is possible thanks to NLP and advanced deep learning models. The structured data allows litigators to see the number of documents, how many they have read, type of the documents, and even historical development.
- Timeline: With access to case files, Uncover also builds a timeline allowing litigators to quickly see which event occurred when. They can also extract all relevant dates from each and every document and also access the data in the form of a chronology of events.
- Draft: Litigators can also draft their arguments while referring to and attaching evidence. This is similar to Google Docs but the only difference being it is for responsible litigation lawyers.
- Search: Another area where Uncover uses deep learning and natural language processing extensively is to power its search feature. This feature allows users to quickly and easily search through all of their files. This step eliminates the need to open documents one by one and scan through the pages for the information you are looking at.
- Collaborate: Last but not the least important thing is collaboration. This tool allows litigators to collaborate with their teams and clients in a single workspace.
Solving a real problem
With Uncover, former litigation lawyers Caroline Korteweg and Ingrid van de Pol-Mensing are solving a real pain point in the industry. The most interesting aspect of this startup is its approach to solve the problem. They are not trying to rewrite any playbook but instead using collaboration as a way to lead the change.
In association with Slimmer AI, they have built a tool that combines core AI technologies like deep learning and natural language processing. With customers like AD Advocaten, CORP., and Lemstra Van der Korst, Uncover is easily one of the startups to keep an eye on in the AI industry, which could become a true disruptor.