Are we entering an artificial world?

Since the turn of the 21st Century technology has progressed at a rapid rate, with many professions using new technologies to adapt, change and improve. We have seen in healthcare technology has slowly been integrated to help with the diagnosis of diseases with machines being trained to help spot specific tumours. In our own profession, one that some from the outside may say is more traditionally rooted and slower to innovate, has been forced to adapt and adopt new technologies as they become available. Through the pandemic lockdown we saw the Land Registry finally taking a step into the modern world (some may say this was long overdue!) with the acceptance of DocuSign, case management systems and subscriptions to legal research websites such as Lexis or Westlaw have now become corner stones of any modern firm.

The latest technological development that is taking up many of the headlines is Artificial Intelligence (AI). We are gradually seeing more and more advanced systems come into play with the most recent AI system catching the attention of the world was Chat GPT. The question, it now seems, is not if but when we will see a more widespread integration of artificial intelligence systems within the legal system.

I recently attended a webinar that looked to discuss this very notion. Whilst this webinar was predominantly focused on the functionality of Chat GPT what was brought to my attention was the fact that legislation is slowly shifting to accommodate the rise in AI usage and how it can be integrated into the legal system. The Data Protection Act contains provisions for AI to ensure that they are processing data fairly, with the ICO (Information Commissioners Office) has identified AI as a ‘priority area’. The Government has produced a policy paper entitled a ‘Pro-innovation approach to AI regulation’ where the focus seems to be to adapting current laws to apply to AI. With the rapid speed at which AI seems to be developing, a more standardised approach may see legislation coming into place that is outdated on arrival. Finally, we can see the EU is putting together an ‘Artificial Intelligence Act’ with a view to try and harmonise the rules on AI.

We can therefore see that the foundations are slowly coming into place for a future where AI is at the forefront of legislation so it is certainly an area in which every lawyer should be looking to keep an eye on.

The question then turns to just how far away are we from seeing a widespread practical implementation of AI? Will people find their co-workers replaced with T-800 terminators within the next 5 years?

A true artificial lawyer is still a distant fantasy at this stage though whilst we may not quite be knocking on the doors of creating Skynet just yet, that is not to say that we will not see a gradual introduction of more AI systems within the next few years. If we look at property law you may not realise it, but we already have AI systems integrated into what we do. ‘Document assembly systems’ are something many lawyers may already be familiar with.

It is this use where I believe we will see the biggest implementation of AI in the near future as it looks to be used as a tool to support lawyers, rather than undermine and supplant. These systems will gradually become more and more advanced, some would say, being able to produce high quality draft documents without the need for numerous amendments and touch-ups on behalf of the human lawyers, although I believe the systems are still a long way off from being able to churn out a perfect lease without any human interaction or supervision whatsoever.

Where I believe the limits of AI lie, and what is really stopping us from having a fully functioning terminator sat on the desk opposite us, is that it can never be able to replicate what I consider to be irreplaceable human attributes. A machine can know all the laws in the world and be able to draft any document one could imagine but if that document contains terms that the party on the other side don’t agree with, it is all but useless in moving a matter closer to completion.  Attributes such as ‘Empathy’, ‘Creativity’, and the art of negotiation are something an AI is very far from being able to replicate. Being able to understand what the other party want, empathising with their expectations, reservations and demands then being able to reach a compromise based on these ideas and the needs of your client is something that an AI is incapable of doing.

‘ROSS’ the AI system first launched in 2016, has been dubbed the ‘worlds first AI lawyer’, however it possesses none of the attributes that I have listed above. Instead it is focused on identifying relevant legal authorities based on questions asked to it in plain English, even being able to ask follow up questions to determine what information is going to be relevant and useful. Whilst it is able to regurgitate information, applying this information in a practical sense still seems to be beyond the capabilities of ROSS. Some scholars, such as Catherine Nunez, don’t believe it’s ‘outside of the realm of possibilities’ to be able to teach ROSS to use professional and moral judgement to handle more complex issues presented to it using the data available, however the discussion of the extent a computer can exercise ‘moral’ judgement is a discussion for another time.

We are clearly seeing the starting foundations of a world built around artificial intelligence, however it is very much still in a fledgling stage. Over the next few years I expect to see a much higher integration of AI systems within firms to help increase productivity, cutting down the more mundane task and assisting in many areas across the profession, ultimately enabling lawyers to spend more time liaising with clients, considering complex legal problems and drafting intricately complicated documents. Both lawyers and legislators should keep a keen eye on the developments in AI as it is sure to have a growing impact on the profession, and everyday life, over the next few years as the technology developers further and further. I do however, believe we are still many years away from finding ourselves sat across the table from a completely independent AI lawyer.