Artificial Intelligence and Construction Law: It’s here, don’t be left behind
Stacy Sinclair, Senior Associate
Artificial intelligence (“AI”) and technological advances are already reshaping the landscape of the UK construction industry. Drones are flying over construction sites creating 3D surveys and robots are building brick walls. This is only the beginning. AI is also starting to extend into construction law. It will not be long before AI and machine learning transform the way in which legal services can assist in the successful outcome of construction projects. The efficiencies and innovations which new technologies can bring to design and construction must now be harnessed by the legal services that support these projects.
There is a lot of talk in the media these days about legal AI and how robots or technology will transform the legal industry and the role of the lawyer. Given the jargon used and the apparent complexities of the technology, it is often difficult to understand what this hype is all about. Indeed, how relevant is it to construction and construction law?
This article looks generally at some of the technologies already available in the legal sector and considers how they might contribute to the success of a construction project or the swift resolution of a construction dispute.
With the assistance of new technologies and more collaboration between construction professionals and their lawyers, is it possible to generate greater efficiencies in the construction process and minimise the existence of disputes?
AI: the jargon
To start with, what is AI? Perhaps Deloitte’s simple definition is most helpful. AI is:
“the theory and development of computer systems able to perform tasks that normally require human intelligence”.1
As journalist and author Joanna Goodman summarises:
“Basically, artificial intelligence is about machines (computer software) doing things that are normally done by people.”2
That sounds relatively straightforward. So what about all of the other terms out there? First, it is useful to know that people often use the term AI generally, when in fact they mean only a small subset of AI or perhaps even a technology that does not employ AI at all. Michael Mills, co-founder and chief strategy officer of Neota Logic, defines the field of AI as having seven branches: machine learning, natural language processing, expert systems, vision, speech, planning, and robotics.3
Others consider that much of the discussion about AI is actually a discussion about pattern recognition within text and the automation of extracting this text. Therefore it is not necessarily pure AI. So terminology and discussions you come across simply may be a particular subset or indeed not AI whatsoever.
In any event, rather than focusing on the specific AI process or technology employed, we first need to consider what the industry needs or wants. What do you want your technology to do? In order to obtain greater efficiencies in each of our disciplines, we need to identify the issue, work stream or “use case” and focus on the outcome or product of AI. Only once we identify the outcome required or the problem to be solved, can we then harness the various platforms/technologies to realise these objectives.
Mills suggests there are five categories of “legal AI”: legal research, expert automation, prediction, contract analytics and e-discovery.4 In other words, in law we see AI used in e-disclosure, contract analysis, case prediction and document automation. The various branches of AI are utilised to do so and the various technologies or platforms are employed for each of these categories.
In terms of contract analytics, “Luminance” and “Kira” are examples of machine learning, contract analysis platforms which can assist with contract review. These platforms intelligently search documents/contracts, extract text and graphically summarise the information. These tools are not a replacement for human analysis, but they can certainly speed up the process and allow for a greater penetration into larger document sets, whilst minimising the risk of human error. Another example is “RAVN”, which was bought by iManage in May 2017. This is also a platform that organises and summarises relevant information from large volumes of unstructured data and documents.
An example of the use of AI in case prediction is the recent challenge between CaseCrunch,5 a UK-based legal tech start-up, and the lawyers of Kennedys, an insurance law firm. CaseCrunch challenged Kennedys’ lawyers to see who could predict with greater accuracy the outcome of a number of financial product mis-selling claims: CaseCrunch or Kennedys. CaseCrunch won. Using predictive algorithms and the modelling of legal issues, CaseCrunch had an accuracy of 87% in predicting the success or failure of a claim. The human lawyers at Kennedys had an accuracy rate of 62%.
Other examples of technology used in law include:
- Robot Lawyer LISA, a “Legal Intelligence Support Assistant” which generates free non-disclosure agreements (NDAs) for business owners and consumers.
- Termi, Helm360’s voice-activated AI assistant for lawyers which interrogates the Thomson Reuters Elite legal practice management system to request billing and other management information.
- Do Not Pay, a free, online bot created by a Stanford University student in 2016 which helps people appeal parking fines.
- Billy Bot, a virtual AI bot programmed to help people find the right barrister or mediator for their legal problem.
- RentersUnion, a chatbot that provides housing advice for Londoners.
The above are but a few examples. There are countless platforms, chatbots and technologies in the legal sector which assist with document analysis, automation and extraction, prediction, research, etc.
AI and construction law
As an industry, we need to consider the possibilities and use cases for AI (to use the term generally). Whilst lawyers use technology internally to automate their own processes and to assist with legal research and document review, what perhaps is most exciting are the possibilities in lawyer/client collaboration.
Given the technology that already exists, answers to the following questions are just around the corner:
- How can we get a quicker and better understanding of our contract obligations, and not just on one project, but across all of our projects?
- Can we manage and automate notifications within our contracts so that obligations can be met successfully and on time, thereby minimising the risk of disputes?
- Can technology assist in the legal review of my contract?
- Can you predict the outcome of my dispute?
- To what extent can I generate legal documents automatically?
To address these issues, good collaboration is needed. You should also remember that every project is different and may have its own unique characteristics. The following are perhaps three possible examples that could be achieved now.
Managing contract obligations: If you enter into hundreds or thousands of contracts per year, managing and maintaining awareness of your contract obligations is paramount. Would a succinct dashboard of the key contract obligations assist in risk management? Where your contracts or subcontracts are agreed largely on your own terms and conditions, with some amendments unique to each contract, are you able to track efficiently the differences between these contracts? The technology available now, if implemented appropriately, can automate the extraction and scheduling of data from large volumes of contracts. With further development, automating notifications for those obligations with time implications may even be possible.
Automating the creation of your contracts: Following on from the above example, if you are the party generating contracts, using your own standard terms and conditions, do you automate this task? Provided the key contract data is inserted, perhaps through a portal bespoke for your company, automating this process may create efficiencies and minimise risks throughout your projects. Amendments to these contracts could be tracked and any proposed amendments could be automatically emailed to your legal team to review quickly.
A legal review of your contracts: How often do you require a legal review of a contract before entering into it? Would you have more contracts reviewed if the process was quicker and cheaper, highlighting only those risks/obligations that are essential to consider? Would a high-level review of the contract be more helpful than no review at all? Lawyers and their clients could work together to develop an automated contract review and extraction process which assists both the lawyers in the legal review and the client in understanding where risks and possible pitfalls may lie. Developing the platform together will allow the client to tailor the process to its needs, whilst benefitting from professional legal services.
The above are but a few examples of possibilities that should be achievable now. The collaboration between lawyers and their clients, with the use of technology and the development of new platforms and processes tailored to each client’s needs, should enable greater efficiency and minimise risks/disputes where possible.
Over the past few years we have seen huge advancements and development in legal AI (including automation/machine learning/etc.). No doubt in 2018 and beyond we shall begin to see applications specifically advancing construction law.
Indeed, with greater collaboration between lawyers and their clients and the harnessing of the technologies, more efficient contract and risk management in construction and the prediction of cases is possible, if not already here. Don’t be left behind.
- 1. David Schatsky, Craig Muraskin and Ragu Gurumurthy (2014), “Demystifying artificial intelligence”, https://dupress.deloitte.com/dup-us-en/focus/cognitive-technologies/what...
- 2. Joanna Goodman (2016), Robots in Law: How Artificial Intelligence is Transforming Legal Services, ARK Group.
- 3. Michael Mills (2016), “Artificial Intelligence in Law: The State of Play 2016” https://www.neotalogic.com/wp-content/uploads/2016/04/Artificial-Intelli...
- 4. Michael Mills, “AI in Law Mindmap”, reproduced in J Goodman (2016), Robots in Law.
- 5. www.case-crunch.com/#challenge