In the latest breakthrough claim for artificial intelligence in law, a firm today announced what it says is the first successful test of predictive coding for document review in a full High Court trial. International firm Berwin Leighton Paisner (BLP) said it had achieved a successful outcome after a 12-day hearing in David Brown v BCA Trading Limited, in which relevant documents were identified with predictive coding.
Predictive coding is a machine-learning technique for identifying relevant documents among very large collections of material. The system is trained by an expert lawyer reviewing a sample of documents, coding the points that would be relevant in disclosure. The artificial intelligence software identifies patterns to the codes and applies them to the complete archive, learning by iteration as it goes along. The initial sample might be 1,000 documents out of a total database of several million, Nick Pryor, BLP’s head of client technology said.
This method of selecting relevant documents is well established in the US but relatively novel in England. It was first approved in early 2016, in a case where both sides agreed to rely on the technology. BLP says the BCA case was the first where the parties disagreed: the firm had to obtain a court order overruling the other party’s opposition to the technology. BLP was representing BCA Marketplace in an unfair prejudice claim brought by a minority shareholder in a business BCA had acquired.
According to the firm, the judgement in the case relied heavily on documents which had been disclosed with predictive coding. ’Seeing the technology successfully deployed in a case and thoroughly tested during a 12-day High Court trial evidences its reliability and makes a compelling case for its use given the cost savings it offers,’ BLP said.
Oliver Glynn-Jones, head of commercial dispute resolution, said: ’We are delighted with the successful outcome of this case, both as a positive outcome for a valued client and as evidence of how predictive coding can be deployed effectively during litigation. Now that the technology has been tested and proven at full trial, and demonstrated benefits in terms of cost and accuracy, we predict that it’s likely to become much more prevalent in commercial litigation.’