top of page
  • Writer's pictureDurham Pro Bono Blog

Automated Lawyers and Robot Judges: How AI-proof is the legal field?

Disclaimer: The views expressed are that of the individual author. All rights are reserved to the original authors of the materials consulted, which are identified in the footnotes below.

By Alex Roper

The issue around the automation of jobs and the resulting obsolescence of human work is perhaps the only concern that is universal amongst aspiring professionals in the new decade. Identifying a role, even an entire field of work, that will survive the employment revolution is arguably as daunting a task as any university graduate will face.

Since the industrial revolution, it has been necessary, in the name of economic advancement and the improvement of standards of living, to replace human labour with automation. The process has already claimed a myriad of employment fields; from the assembly line to the supermarket checkout, the impact of technological progress has been felt by workers and businesses alike.

This process has accelerated exponentially during the 21st century. According to a 2017 analysis of UK employment by the Office for National Statistics, 1.5 million English workers are at ‘high risk of losing their jobs to automation.’[1]

Many university undergraduates now wisely choose their degrees with this question at the forefront of their minds. This perhaps makes enrolment in Law school an attractive option: almost 28,000 students applied for Law degrees in the UK in 2018-19.[2] Many of these students aspire to become solicitors, barristers, legal counsels to corporate firms; some will hold hopes to enter the judiciary as judges in years to come.

But the question of whether the legal system is as immune from technological advancement as many would believe is a complicated one. Could the field be revolutionised by artificial intelligence and machine learning?

Artificial intelligence (‘AI’) refers to computer software systems that have the ability to ‘learn, plan, reason’[3]rather than being reliant on pre-programming.[4] This article will consider a number of legal roles that could be affected by the development of AI, and attempt to address the overarching issue of the legal profession’s safety from technological progression.

Paralegals and legal researchers

The roles of legal researcher and paralegal are the most susceptible to artificially intelligent software. As is the case in all fields, jobs that require less training, and involve more repetitive tasks (particularly when these tasks demand the processing of huge amounts of information) are at increased risk of being superseded.

There are currently already AI algorithms that are more efficient than humans in this role,[5] despite the use of the technology in the field still being in its infancy. A 2019 Forbes article predicts that AI will be disastrous for legal researchers.[6]

However, for those willing to adapt to technological advancements, AI could be used collaboratively as a tool to work more efficiently and increase output. Whilst the number of jobs in the field will clearly be reduced, those remaining will be able to ‘focus on higher-value tasks,’[7] taking up a more technical role within firms;[8] 63% of paralegals already conduct e-discovery work,[9] showing clear potential for lower-level law firm employees to become technological intermediaries between AI applications and lawyers.


A highly skilled job in the legal field, the role of a solicitor is far better protected against the developments of AI.

Nevertheless, the idea that solicitors are immune from AI advancement is something of a fallacy. Research suggests that 23% of a solicitor’s tasks can already be automated with existing technology;[10] AI can aid (although this naturally leads to some level of replacement of lower-level roles) solicitors in ‘electronic discovery, due diligence and contract review,’[11] presenting an exciting opportunity for firms looking to meet the needs of their clients. The consensus in the legal field, currently, is that AI will become a vital tool to increase efficiency and allow for the automation of more monotonous tasks, whilst solicitors focus on higher-level issues and interacting with clients. A paper from the

University of North Carolina School of Law[12] suggests that even using all the current technology could decrease hours worked by solicitors by 13%, a surprising estimate given the current practical and technological limitations of AI technology, and an indication of potential upheaval even within the higher levels of the profession.


Barristers would be incredibly difficult to replace with AI, given that the majority of their workload is less focussed on repetitive tasks and the processing of information.

The role is perhaps safer from automation and AI because barristers are required to interact with clients, solicitors and witnesses; advocate in court; present arguments (often to juries); and negotiate settlements.[13] The key aspect that ties these tasks together is the necessity of human connection, which is difficult, perhaps impossible, to replicate with AI.


The issue of implementing artificially intelligent judges centres more around ethical and philosophical, rather than practical, implications. The idea seems somewhat far-fetched, but current difficulties within the UK judicial system, including predictions of large ‘backlogs’[14] of cases that courts do not have the capacity to process, could make a technological revolution an attractive concept.

Whilst the UK might have shown little indication of moving towards AI adjudication, the Estonian Ministry of Justice is currently considering applying AI technology in small claims courts,[15] and researchers at the University of Alberta are developing AI that can predict the decisions of judges by analysing evidence.[16] This shows a clear trend towards allowing AI to play a role in adjudication.

The introduction of such technology would be beneficial to some extent. It is undeniable that human judges make decisions based on personal experience, which can lead to cases being influenced by bias (conscious or unconscious). AI, on the other hand, would arguably be more impartial, relying solely on the data.

However, studies suggest that current technology may not be objective. COMPAS, a machine learning algorithm, was found to incorrectly estimate the chance of reoffending in the US-based on the race of offenders,[17] an example of machine bias that suggests even more advanced AI could be susceptible to bias if applied incorrectly.

The most viable option would therefore be to use judges and AI together, making use of a judge’s empathy and experience, and the AI’s abilities to both counteract bias using data and improve consistency.[18]


It is highly likely that AI will have an enormous, arguably revolutionary impact on the lower levels of the legal profession in the next decade, and will increasingly influence even higher levels as the technology improves beyond this time period. However, the effect will be limited to some extent by the law’s innate need for empathy and human connection, which will always be necessary alongside the use of artificial intelligence.



[1] ‘Which occupations are at highest risk of being automated?’ (ONS, March 25 2019)


[2] ‘Undergraduates and graduates in law’ (Law Society, November 25 2020)


[3] ‘The Power of Artificial Intelligence in Legal Research’ (Lexisnexis, October 9 2020)


[4] Ibid

[5] Seda Fabian, ‘Artificial Intelligence and the Law: Will Judges Run on Punch Cards?’ [2020] 16 Common Law Review 4, 5

[6] Neil Sahota, ‘Will A.I. Put Lawyers Out of Business?’ (Forbes, February 9 2019)


[7] ‘A Paralegal’s Guide to AI and Automation’ (Lexisnexis, March 12 2020)


[8] Ibid

[9] Ibid

[10] Steven Lohr, ‘A.I. Is Doing Legal Work. But It Won’t Replace Lawyers, Yet.’ The New York Times (New York, March 19 2017)


[11] Ibid

[12] Dana Remus and Frank S. Levy, ‘Can Robots Be Lawyers? Computers, Lawyers and the Practice of Law’ (SSRN, November 27 2016)

[13] ‘Differences between a lawyer, a solicitor and a barrister’ (Slater Gordon, September 23 2016)


[14] Nick Davies, Thomas Pope and Benoit Guerin, ‘The Criminal Justice System’ (Institute for Government, April 28 2020)


[15] Fabian (n 5) 4

[16] Fabian (n 5) 5

[17] Ibid

[18] Fabian (n 5) 6

156 views0 comments

Recent Posts

See All

The controversy surrounding facial recognition

By Holly Downes Facial recognition is a technology that has maximised our safety and protection, served justice by solving crime cases, and has made identification safe through authentication, yet the


bottom of page