Free Newsletter Free Newsletter

BLP wins in disputed UK predictive coding case David Brown v BCA

Added on the 12th Mar 2018 at 2:02 pm
Share Button

Following the first English contested court order for the use of predictive coding in the document review process, the litigation and corporate risk team at Berwin Leighton Paisner (BLP) have achieved a successful judgment at the end of a 12-day High Court trial in the case of David Brown v BCA Trading Limited.

BLP, who acted for BCA, says that this is the first time that predictive coding has been tested through to full trial – it is certainly the first time following a contested application.

In May 2016, BLP secured a landmark order to use predictive coding in the face of opposition from the other side. This set it apart from the similarly landmark UK judgment in the case of Pyrrho Investments, where the parties were in agreement about the need to use predictive coding in the document review process.

The BCA team at BLP was led by litigation and corporate risk partner Oliver Glynn-Jones, who told Legal IT Insider: “At the case management conference we had to make the case for predictive coding and we presented evidence to show that it is better than human review and around a third cheaper than manual review. All the safeguards the Court was looking for were in place and we persuaded the Court that it was the right way to proceed.”

He added: “This is a case where the documents were key and pivotal to the judgment – and that’s what came out of the predictive coding exercise.”

Predictive Coding is machine learning technology that has the potential to dramatically reduce the cost of the e-disclosure process.  A senior lawyer reviews a small “seed set” of documents, which is then analysed by the technology and used to generate a further sample for review.  Through a process of iterative refinement, the algorithm can reach a level of review accuracy that can be applied across the entire dataset, identifying relevant documents in a manner that is far more efficient and scalable than a traditional document review.

The success of BCA can be expected to pave the way for more trials using predictive coding technology. Glynn-Jones said: “The reality is that the courts are already aware of the technology and want to use it. If you have an example where it’s been used in a contested manner and it’s been successful, then people will point to it as a demonstration of it working in practice.”

While Glynn-Jones says it’s difficult to be definitive over whether the exercise was a third cheaper, he adds: “I’m pretty confident that from the initial disclosure done versus what would have been the costs, there’s a very significant saving.”

While machine learning technology is now widely used by the largest UK and global law firms in the corporate due diligence process, the contentious nature of litigation and the concerns around missing case-critical documents in the review process have meant predictive coding has not made the same progress.

However, Glynn-Jones said: “Everyone’s focus is always on the cost of predictive coding and people ignore the fact that the evidence shows that it is more accurate than human review.

“People are concerned that a computer might miss key documents but they will come to realise that it’s all about the initial work you do as humans teaching the algorithm: as long as that initial training is of fantastic quality, the machine will deliver better result than paralegals at the end of eight to 10 hours of reviewing documents.”

BLP is using a number of different artificial intelligence-based technologies including iManage RAVN’s Applied Cognitive Engine (ACE). It was also the first UK-based law firm to use Opus 2’s Magnum platform for unilateral case management.

Glynn-Jones said in a statement out today (12 March): “BLP prides itself on being at the cutting edge of using law tech and AI to improve the delivery of legal services, and we are almost unique in having the predictive coding platform and forensic disclosure specialists in-house. Using this setup, BLP was able to significantly reduce our client’s costs. 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.”

In addition to Oliver Glynn-Jones, the BLP team included Robin Ganguly, Rebecca Wardle and Alasdair McAlpine. The team instructed Sa’ad Hossain QC and Joyce Arnold of One Essex Court.


One Comment

  1. Chuck Henrich says:

    One of the biggest risks in relying on assisted/artificial intelligence or predictive coding lurks in the prejudices and “unknown unknowns” embedded in the algorithms doing the search and evaluation. Humans (lawyers) can think outside the box, computers can’t (literally). Aside from the inevitable limitations of human-generated algorithms, training the machine will also involve mistakes and limited vision, because people are fallible. Key quote from this article: “… *as long as that initial training is of fantastic quality*, the machine will deliver better result than paralegals at the end of eight to 10 hours of reviewing documents.” (emphasis added) Unfortunately “fantastic quality” is by definition rare. Usually when humans are involved, quality is somewhat middling, if not poor. So beware blind faith in technology.

Any Comment?