It does not take an observant lawyer to spot that there are a growing number of companies entering the legal sector with new ideas about how technology can be used to improve service quality whilst simultaneously reducing costs. They are bringing a wind of change that is hard to resist for lawyers, whether practicing in law firms or within inhouse legal teams. Automation, driven by Artificial Intelligence, in particular, is creating a wide range of improvements to legal processes. However, the sector needs to be careful in how it goes about adopting AI as law is unlike most other industries when it comes to data and how it should be used.
In the new data goldrush it is important that legal professionals should not simply copy how other industries have used technology, but to think through carefully about their unique situation to realise the full benefits.
The first challenge is Big Data, or rather the lack of it for law. Other sectors have been building up huge reservoirs of data over many years, creating the environment for AI to thrive. Retail, for example, is one of the first industries to fully embrace predictive technologies. Numerous types of data are collected when consumers interact with a retailer, such as what demographic they belong to and what they’re likely to buy alongside a certain product. These conclusions are reached through multiple data sets, where machine learning and AI leverage information to predict a set of outcomes.
However, in the legal sector, data is harder to quantify. Traditionally handled in an analogue, face-to-face format, data has not been stored in a curated way that allows for the interrogation by an algorithm or AI tool. However, this is changing. Most lawyers use dynamic templates, which construct an initial document from answers to a questionnaire, which the lawyer then tailors as appropriate. With law firms holding significant quantities of data from this process, including documentation from past deals and transactions, email correspondence and phone records, data can now be curated in a structured way that allows for the building of Big Data. Using such tools as hybrid reasoning software, law firms will soon be able to potentially build a collection of big data and then leverage machine learning and AI tools to tap into what has traditionally been largely unstructured records,
A second difference is concerning specificity. Unlike retail, where AI is used as a predictor of consumer behaviour, the legal industry cannot afford to rely on anything other than specific advice. If lawyers are using technology to reach a decision, then the answer needs to be precise, there is no room for error. It’s not possible to rely on past situations, as advice has to be specific to the exact current set of circumstances. It is, therefore, harder to experiment with the use of AI in the real world – the legal sector needs to get AI right the first time. After all, the ramifications of incorrectly predicting that someone would like to buy a pair of pink socks with their shirt purchase is on a wholly different level to advising on the structure of an employment contract or other legal documents.
A third difference is the cultural issue in law of the relationship lawyers have with their clients. There is a strong sense of personal human contact between the lawyer and the client as the basis for how the sector operates. The lawyer spends a lot of time (and money) building up their knowledge and experience, building a reputation that attracts clients. However, the changing world demands that this mindset needs to be challenged in certain areas of legal activity. The capabilities of AI and automation in particular means that large amounts of routine tasks can be completed in more cost-effective ways. This trend of automation should not be resisted, but embraced, with legal industry business models being adjusted accordingly. Lawyers who do not adjust will quickly find themselves being outmanoeuvred by more effective competitors. The knowledge/experience/reputation equation is still valid in many areas of law but this will be increasingly at the more complex end of the spectrum. For the everyday, a good algorithm and user-friendly online interface will be cheaper, quicker and just more convenient for people to use.