Earlier this year, the Government announced an increase in LawtechUK’s funding to drive economic growth through greater adoption of AI and technology in legal services. Private sector investment continues to ramp up, with recent funding rounds valuing Harvey and Clio at $3bn apiece.

Through its adaptability, generative AI has enabled tech companies to rapidly deliver impressive functionality that would have previously been uneconomical. 

As a result, more and more lawyers are beginning to use generative AI for their legal work. However, legal work is complex and requires a consistently high level of accuracy. Lawyers must now learn how to optimise their inputs and carve out time to review.

What’s design got to do with it?

Using generative AI for legal tasks creates a tension between accuracy and speed. Initially, generative AI products required users to experiment with prompt engineering and review outputs with a fine toothcomb.

For lawyers, this is not a satisfactory experience. Some may prefer to return to a manual process to ensure accuracy whereas others might truncate the review process to realise time benefit.

Recently, a US attorney representing an AI company in an ongoing copyright dispute admitted a citation error created by a generative AI tool. Whilst the link, volume, page number and publication year were accurate, the author and title were hallucinated. The attorney had to explain to the court that the citation error was not picked up during their manual citation check.

It would be easy to dismiss this as an example of human carelessness; the attorney should have been more rigorous in their review or not have used generative AI in the first place. 

A user experience (UX) designer might come to a different conclusion. Instead of blaming the user, the interface of generative AI tool should have been designed to empower the attorney to easily identify and correct the error. 

In short, UX designers seek to address these challenged through the application of straightforward design principles that bridge the gap between humans and technology.

What is UX?

UX is an intriguing blend of technology, psychology, and design. Its goal is to harness our understanding of human behaviour and cognitive biases to create digital processes that help users achieve their objectives with minimal friction or assumed knowledge.

Several psychological principles underpin effective UX design. The “aesthetic-usability effect”, for example, suggests that users perceive attractive systems as more usable. “Chunking” – breaking up information into manageable pieces that the human brain can process – can help to overcome the decision paralysis caused by choice overload. Think about how you remember your phone number; not as a single string of digits but as bite-sized pieces. 

In the 1980s, studies saw that when computer systems respond quickly (ideally within 400 milliseconds) users become more engaged and productive. Where longer processing times are unavoidable, designers developed loading animations and progress bars to make waiting more tolerable. 

In today’s digital world, users bring expectations shaped by their experiences with other technologies. UX designers can leverage these familiar patterns to help users focus on their tasks, rather than learning new processes from scratch. 

A digital tool that is responsive, uses familiar inputs, and presents information clearly can help users achieve a state of “flow”; that sweet spot where focus is maximised. 

Why good UX is essential for technology and GenAI in the legal sector

For UX designers, form is function.

If lawyers are to embrace generative AI, the technologies must be designed with their needs at the forefront. Surprisingly, this is sometimes overlooked in the race to develop ever more advanced features.

Generative AI products that struggle to facilitate these steps risk alienating users and undermining promised efficiency gains. This can lead to lawyers turning back to more familiar and manual processes.

By understanding how the attorney was using the generative AI tool, a UX designer would clearly see the risks associated with submitting erroneous citations within court documents. Accepting that hallucinations do occur when using generative AI, the interface would need to be designed to assist the attorney to verify sources efficiently and reduce the risk of errors slipping through the net.

As we see greater investment in the legal technology industry, we should expect to see further developments in the UX design of generative AI products geared to how lawyers operate and can be responsive to certain areas of specialisation and expertise. 

Ultimately, UX designers are concerned with building a harmonious relationship between human users and technology. When UX principles are embedded in a solution’s architecture, users and machines can work together seamlessly, resulting in more reliable outputs, greater accessibility, and increased adoption.