Specialists have spent the last decade building innovative, engaging, and engaged AI products for a wide range of companies. Retresco for example currently has NLU and Natural Language Generation (NLG) projects running in the media, finance, sports, and weather sectors, transforming text into data on some projects and data into text in others.
NLU and NLG are based on the same technology and people working in both departments share knowledge, experience, and developments. By understanding and improving one part of the technology, they can understand and improve the other. NLU is used to analyse and assess user input. Once this is analysed, the NLG technology steps in to supply the correct answer. NLG is the process of taking data and turning it into readable, publishable text. It is often referred to as ‘robot journalism’, ‘automated content’, or something similar.
In general, the technology can be used in any sector that has a mass of well-structured data and is confined within strict boundaries and regulations. The lawyers will not be replaced, but will instead be supported in routine work such as the checking and creating of standardised documents. This enables them to be more dynamic in their reactions and to be able to apply more time to individual queries and issues. This brings several immediate benefits for law firms, chiefly that it saves lawyers and their staff valuable time. This frees them to do the more-creative, specialised work that relies on information from beyond the data pool and is the type of task that is outside the limits of computers. There are also the benefits of speed and accuracy, the former since computers work much faster and continually—unlike humans—and the latter because machines do not make mistakes. NLG has benefits that extend beyond the legal profession. For clients, it means a more individual and targeted campaign.
Another focus could be the implementation of chatbots. These, too, have applications within the legal sector, offering a significantly faster and solution-oriented support for clients. Initial information can be gleaned and classified, before the case is handed to a lawyer. While some situations are not suitable to be handled by a chatbot but they can be applied when it comes to standardised business models. For lawyers, themselves, chatbots could be applied for giving background information on specific laws and regulations. A good example of this may be the General Data Protection Regulation, which came into force in 2018.
We are already seeing AI being applied within the legal sector, and the indications are that this is a trend that will continue to grow. Statistics from www.job-futuromat.iab.de indicate that 71 per cent of paralegal activities today can be automated. For the activities of notaries, this stands at 67 per cent. We also see, across the industry, values of 50 per cent for commercial lawyers and solicitors, 33 per cent for specialist solicitors and patent lawyers, and 20 per cent for attorneys-at-law. The future is here, with more on the way.
It takes time for an industry to adopt change, and to allow those changes to bed in. It is akin, in many respects, to a tanker changing direction. Wholesale change within an active sector is difficult to implement. The law is continually shifting and moving, and any implementation needs to be integrated seamlessly within an already-moving machine. This is something that needs to be managed well.
Another challenge is the redistribution of tasks and roles for support staff within legal firms. Successful lawyers usually have a bevy of support staff that help them to manage their processes and work habits. Implementing AI into this milieu will require management of those members of staff in their transition to roles and tasks that are not possible to be done by AI.
There are two trends within the legal industry. Firstly, while AI is not a match for – and cannot replace – a flesh-and-blood lawyer, the back-office functions in any legal firm will benefit from the streamlining that AI brings. Implementation of many of these tasks is underway and is more palatable for those in the profession, being that they aid employees but do not replace them. The second area in which we foresee growth is in start-up companies such as Flightright, which operate on business models that excise the need for a professional lawyer. In that company’s case, legal applications and matters are mostly handled online, using information input by the plaintiffs themselves. Start-up companies like these not only employ new business models but are highly efficient, focus on value, look to uncover new customer segments, and are driven by technology. It is in these spaces that we think there is great potential for growth.
But while there are practical concerns, we find from our experience that the main challenge in implementing NLG content in any industry has often been an inner recalcitrance and fear of change. There is a justifiable wariness that arises when it is perceived that interlopers are making claims and promises about an industry that they may have little direct experience in.
This, however, remains our strength. In developing systems from scratch, Retresco brings a fresh pair of eyes and a willingness and ability to understand complicated subjects. A decade’s worth of experience gives us the real ability to not only implement an NLU or NLG system, but to ensure it is embedded properly within the workings of a company.