Global law firm Berwin Leighton Paisner (BLP) is in discussions with a specialist provider of artificial intelligence (AI) computing to improve its ability to search documents in the field of commercial contracts.
The move parallels a growing awareness of the potential of robotic process automation (RPA) to automate process workflow routines by large commercial businesses. The Co-op Bank is reportedly using RPA to make debt collection decisions and Co-op Energy uses it to process thousands of new customer applications.
While not AI in the futuristic sense of self-aware robots, or even machines that can make sense of unstructured information, the term RPA is used to describe sophisticated process robots built using computer code, similar to the ‘agents’ that were previously common in the world of process automation. But rather than have to be written laboriously to perform narrow tasks, today’s RPA robots can read, write and report data across different systems within an organisation.
BLP’s head of legal risk consultancy, Matthew Whalley, who helps clients identify and mitigate legal risks, told Legal Futures that the next big advance in AI for the legal market was not just searching and finding a list of relevant documents, but extracting the information and data within documents.
He said: “If you can get a bit of software, whether we call it a robot or a routine, that can index through 10,000 documents, identify the 100 documents that are relevant… and identify five data points from each document… actually what you’re talking about is a very advanced and quite a revolutionary software routine.
“That core ability to extract and report data from documents is the key thing that has actually been missing from search technology and review technology for the last 20 years.”
He said this use of AI was “on the near-term spectrum”. While much discussion of AI was what might happen in 20 years’ time, “for now the real challenge is that ability to extract and report on key data points and it’s really hard to do, especially on a large set of unstructured data, which is what a big body of contracts is.”
He added that RPA and potential associated AI was “quite an exciting technology, particularly if you are in the legal industry and particularly if you are involved in legal risk analytics and contract risk analytics, and you want to find solutions to sifting through hundreds of thousands of documents and reporting on the things that actually matter without involving thousands of man-hours”.
Mr Whalley concluded: “As RPA advances, it has a role in any organisation that has repeatable processes, that employs people to do repeat work. If you invest the time in training, or encoding your robot… [it] will work tirelessly throughout the night, every night, 365 days of the year and never deviate or make any mistakes. It means you can free up your people to move up the value chain rather than just trap them in boring, repetitive, menial tasks.”
He confirmed that BLP was negotiating for the services of a firm “that delivers AI, or really good search, in contract governance and analytics”.
At the end of last year, BLP built its first internal contracts robot and said it expected to build more robots for clients in the coming years, which would cover tasks such as processing contracts, monitoring payments against contractual terms, and improving the accuracy of standard contracting negotiations.
Earlier this month, we reported that Riverview Law had entered a partnership with Liverpool University to apply computer science expertise in AI to legal services. Riverview’s chief executive, Karl Chapman, is also chairing a technology advisory council set up to work with IBM and others to shape the development into the professions of ‘cognitive’ decision tools such as IBM Watson. Watson claims to be able to find “answers and insights locked away in volumes of data”.
A report last year by Jomati predicted that robots and AI will dominate legal practice within 15 years, perhaps leading to the “structural collapse” of law firms.
i agree this is close, if not already ere. Check out diligenceengine for example.