Technology has frequently hit the headlines recently – for good and bad reasons – and it is clear that automation, computerisation and artificial intelligence are marching forwards with huge implications for how we live and work in the future. No job will be immune to technological development, and some will be changed beyond recognition or even die out. When SIArb decided to organise an event aimed at younger practitioners, a consideration of how these trends will affect dispute resolution seemed the ideal topic: since younger practitioners will be most affected by these changes. And so, on 6 December 2018, 110 delegates (young and old) gathered to hear leading younger members of the arbitration community debate these issues.
Proposing the motion “This House Believes That Artificial Intelligence Will Have Replaced Arbitrators Within 25 Years” were Divyesh Menon (Mayer Brown) and Katie Chung (Norton Rose Fulbright). Divyesh and Katie argued that the use of artificial intelligence to resolve disputes could eliminate irrationality and bias, boost the predictability of decision-making, aid finality, and save time and costs. Scientific research, they said, suggests that the professional intuition and instinct which lawyers employ is really an example of pattern recognition by the brain. Artificial intelligence can analyse documents and recognize the same patterns more quickly and more reliably than humans, and bring that to bear to decide disputes. Research has shown that artificial intelligence can already predict decisions of the European Court of Human Rights with a high degree of accuracy, and it will only improve in the future. Ignoring this, Katie said, was to act like an ostrich with its head in the sand.
In response, Patrick Dahm (Samuel Seow Law Corporation) and Benson Lim (Hogan Lovells) countered that artificial intelligence does and will always lack certain things that humans possess: consciousness and empathy. These features, not irrationality, are what human beings really bring to the arbitration process and they are important to users. Pattern recognition and other technology will, no doubt, be used to help arbitrators do their job better but will not replace them. Even if some disputes are resolved by artificial intelligence alone, that will not be acceptable to many parties. Furthermore, just as robots moving around a field and pushing a ball is not football, dispute resolution by artificial intelligence would not be arbitration.
Judges Ankit Goyal (Allen & Gledhill), Alessa Pang (Rajah & Tann) and Jae Hee Suh (Allen & Overy) probed the debaters’ arguments. Would artificial intelligence really eliminate bias, asked Jae Hee: who would develop the algorithms and determine the data sets to be analysed? Alessa was sceptical of the claims of greater finality. The verdict reflected this: the judges came down against the motion, saying there would still be a role for arbitrators in 25 years’ time. Would it have affected their decision if the motion asked whether artificial intelligence would replace arbitrators in 50% of cases (asked moderator Sapna Jhangiani of Clyde & Co)? It would have made the decision much more difficult, confessed Ankit, conceding that there could be a decision-making role for AI in some future cases.
The audience also had a chance to ask questions and vote. A show of hands at the start of the event showed the proposers faced an uphill struggle: only 6 people were inclined to agree with them, and a similar number were undecided. By the end they had changed some minds, gathering 21 votes, but it was not enough, and the motion was defeated. All would agree, however, that the eloquent arguments had provided much food for thought, and many of the audience and speakers continued the debate over the networking drinks which rounded off the evening.
CLICK HERE to view more photos.
Andrew Pullen – Council Member