Legal Considerations In The Use Of Artificial Intelligence

Marcus-Okoko & Co


Marcus-Okoko & Co logo
Marcus-Okoko & Co is a full-service International law firm committed to providing a wide variety of topnotch legal services to clients within and outside Nigeria. Our areas of expertise cover a broad range of corporate and commercial legal services amongst which include Corporate and Commercial Law and practice; Arbitration; Litigation; and other forms of ADR such as Legal and Regulatory Compliance; Mergers & Acquisitions; Competition & Antitrust; Project Management and Capital Markets.
Technology, as a concept, and a practical application of knowledge, has pervaded every given sector of life, including the Legal Practice. A core subset of Technology that is fast gaining...
Nigeria Technology
To print this article, all you need is to be registered or login on


Technology, as a concept, and a practical application of knowledge, has pervaded every given sector of life, including the Legal Practice. A core subset of Technology that is fast gaining a high level of infusion into the Legal practice is the concept of Artificial Intelligence. Artificial Intelligence (AI), like other sectors and spheres of influence is beginning to see its relevance being felt by Lawyers and AI engineers alike within the legal industry.

Traditional tasks which can be handled by Lawyers have been seen to be handled by Artificial Intelligence, including but not limited to Data Processing, Legal Research, Case management, and case outcome prediction. In capacity, Legal Tech continues to advance and scale borders every year.

With the adoption of Data Science, Deep Learning, and Machine Learning algorithms, the disruption of AI has affected different industries. Rather than a threat, AI is an innovation. Due to the seeming conservative nature of Legal systems of developing countries, these developing countries seem largely unprepared for the transformation which Artificial Intelligence will cause in the Legal sector

In this paper, a deeper insight into the concept of Artificial Intelligence will be given, alongside the convergence of AI and the Law, and the ethical considerations to be taken note of along the lines of Artificial Intelligence.


Artificial Intelligence (AI) has been defined by Investopedia1 as the simulation of human intelligence machines which are programmed to think like humans, solve problems like humans, and equally mimic their actions.

Artificial intelligence, through the concept of knowledge engineering, is based on the fact that human intelligence can be outlined in a way that a machine can mimic it, and execute tasks more efficiently, whether the tasks are simple or complex. These tasks range from activities like learning, reasoning and perception.

Worthy of note is the fact that giant strides of development have been achieved along the lines of technology and artificial intelligence. For example, machines which calculate very basic functions were at some point regarded as artificial intelligence. Now, however, these kinds of functions are no longer regarded as Artificial Intelligence, but are now seen as inherent computer functions which involve very simple algorithms. Artificial Intelligence is currently being applied to diverse industries and spheres of influence. In the healthcare industry for example, AI is used for dosing drugs2, and surgical procedures3 in the operating rooms of hospitals. There are equally machines that play chess excellently4, and there are AI powered self-driving cars5. The Financial industry has also got Artificial intelligence applications6 which are usually used to detect suspicious activities in finance, including but not limited to unusual debit card usage and large account deposits. These applications help a bank track the occurrence of fraud, and helps out with the different Know-Your-Customer (KYC) requirements which different banks have. They equally also help in making bank trades easier in the realms of stocks and Forex.


There are several ways through which the copious functionalities of Artificial Intelligence can be infused into the practice of Law.


Ordinarily, the traditional Legal practice is laden with a whole lot of paperwork. There are lots of documents which facts have to be extracted from, and which have to be reviewed in order to give the appropriate legal advice in that context, or to spot errors or documentary booby traps. The traditional approach of reviewing documents on a page-to-page basis by humans is vulnerable to human errors. Time which can be spent on more profitable ventures is usually wasted on humanly reviewing documents.

The usage of Artificial Intelligence in document review has proven to be far more efficient. The review of documents through Artificial Intelligence is usually very quicker, and more efficient. It possesses the capability7 of identifying differences between contracts in a due diligence process, and ensures that key provisions are highlighted. Relevant documents are identified and analyzed, while noting the key provisions which are to be looked out for. Bech-Brun8and ROSS Intelligence9 are examples of companies that offer document review services. It however should be noted that today's AI is not so strong, and it would still need the supervision of human lawyers.


In the corporate sector, due diligence is a very important role that lawyers play for different reasons. Due to the versatile skills which Lawyers possess, new generation lawyers may be asked to conduct due diligence on tech startups or tech companies by Individual or institutional investors. Also, along the lines of complex transactions like Mergers & Acquisition, due diligence is needed.

Due diligence is a tedious task. It could involve the review of many documents and contracts. It involves the collection of diverse records that are in different frameworks and different places. This process costs a lot of energy and time on the part of Lawyers. This review process could involve preparation, setting given objectives and categorizing financial information. A mistake made while conducting due diligence could affect the entire firm. An example of a bad due diligence situation was exemplified in Hewlett Packard's deal with Autonomy10 where an impairment charge of 8.8 billion was recorded by the acquiring company.

The Adoption of artificial intelligence in due diligence processes is advantageous to Lawyers in diverse ways, as listed below.

  1. It saves the time and effort of Lawyers conducting due diligence. Artificial Intelligence enables a due diligence job to be completed in a short time with high accuracy.
  2. Artificial intelligence can also be used to identify all the relevant documents which are to be used for transactions relating to Mergers & Acquisition, and generally, due diligence. The AI can be programmed to familiarize itself with diverse legal provisions which are seen in the due diligence documents, including but not limited to: dispute resolution, jurisdiction, anti-dilution clauses, indemnification, non-competition, non-disparagement, etc. AI can review documents which seemingly look alike, and can pinpoint the differences in the documents.
  3. Artificial intelligence can effect the identification, classification, organization and prioritization of documents. It can equally determine the exact documents that need to be provided by the company which due diligence is being done on.
  4. Artificial Intelligence can determine key issues to look out for while conducting due diligence, highlight and catalog the relevant provisions of each document, while Lawyers can easily look out for, and sort out problematic sections, if any.


For Lawyers in the litigation sector of legal practice, Artificial intelligence-based tools can help predict outcomes of court-based cases by analyzing a wide expanse of historical judgments, looking at the peculiar nature of the case and looking at the decisions which have been reached by the Judge. Trends can also be taken note of and spotted by the AI prognostication tool, and these could be used to influence the prediction of the AI case. This prognostication technology might not exactly predict strategies (at least for now), but they can provide an unbiased prediction of how a case will turn out eventually.

Lex Machina11 and Solomonic12 are examples of Legal-tech based prognostication technologies as they predict likely outcomes of substantive suits by sifting and reviewing a lot of court judgements, and likely outcomes based on past data. The information which is gotten from the AI tool helps lawyers and arbitrators alike prepare for their cases, and reduce risks or strategies that will not be of advantage in their cases.


In some areas of Legal practice like Arbitration and Litigation (even some parts of corporate practice), there is the issue of divided attention because of repetitive administrative functions. AI-powered systems can perform repetitive administrative tasks through self-learning capabilities. These systems can very well learn from past events, and improve on their subsequent outputs.

In Arbitration, although parties enjoy party autonomy, which enables them to make their choice of arbitrators, some AI Software can help parties to a dispute make the best of decisions in their choice of arbitrators. These decisions would be based on the existing track records and the specialty of the arbitrators.AI powered legal-tech Software can streamline and take care of a number of administrative tasks, thereby leaving Lawyers to focus more on the parts of the dispute resolution process which needs a high amount of human judgment. For very voluminous documents, a software like Disclosure management13 uses predictive coding and natural language processes to extract the gist of a document. Earlier, Disclosure Management could only search and extract keywords, but now, it can explain and convey the actual meanings embedded in documents.

Administrative issues can be easily handled with the use of automated case management software. All documents which are to be relied on in a case can be run exclusively through a single platform, especially in commercial arbitration, which thrives on confidentiality. AI-powered case management software can be programmed to highlight references to exhibits that are to be relied on in the course of the substantive suit. Hyperlinks can be used to reference specific sections in Laws of the particular jurisdiction where the case is being held. All these help judges and arbitrators alike to follow the link of argument of the parties to the suit. This renders the need for a lot of hard copies of documents and exhibits useless. In the realms of arbitration, the concept of case scheduling can be initiated by the platform. Arbitrators, counsel and parties could indicate times which they would be available for hearing. The case scheduling software would compile a list of hearing dates that would suit everyone.


Artificial Intelligence has gained a high-level ubiquity in different sectors, and the Legal practices is not left out. The ubiquity being talked about is largely as a result of the mind-blowing advantages, and the potential for limitless possibilities. Artificial Intelligence however has some potential for negative effects, and thus raises the need for ethical considerations in areas such as data protection, human rights, Torts (liability) and Intellectual property.

As intelligent as AI systems are, the concept of 'Garbage in, Garbage out' still exists and is still largely operational. AI systems are not always capable of giving the best of legal advice as they act based on the data fed to them, and the level of their self-learning capacities. The present forms of Artificial Intelligence systems available can at best mimic emotions, but are not capable of having emotions in themselves. Without emotive measures in place, the advice which will be given by the AI systems may not be in the best interest of the client. Where parties to a suit or clients in the corporate world suffer because of the half-baked advice of AI-powered systems, the Lawyers, and not the AI system, would bear liability based on negligence. In cases where documentary evidence would not suffice to cover the entire sphere of the dispute (which is usually the case), the advice given by AI-powered systems cannot be trusted. Disputes with legal complexities and larger-than-life issues need handling by humanoid practitioners. AI systems can at best assist.

Debates as regards privacy safeguards and how to overcome bias in algorithmic decision-making are now familiar14. It is highly probable that consciously and unconsciously, program developers and software engineers have built prejudice and bias into the datasets used to train the software.

In the Legal practice, the data of clients which Lawyers represent are usually of a confidential nature. In various jurisdictions, Lawyers owe their clients a duty to be confidential (attorney-client privilege). If these data are being processed by AI systems and machine learning tools, Lawyers will be held responsible if there is a breach of the data involved, or if there is non-compliance with data protection laws. Legal practitioners that to make sure they exercise control over the actions of these AI-powered systems to make sure that confidential information of clients is stored safely


Although Artificial Intelligence is not intended to be an apocalypse for Lawyers and the legal profession, the next few years will see a revolutionary change in terms of the Law, and how it is practiced, enforced by the executive arm of Governments, and interpreted by the Judiciary. AI practices and tools will largely become ubiquitous in the legal profession.

The application of AI in legal practice will no doubt revolutionize and develop the legal industry, it is however pertinent that the ethical concerns be taken seriously by Lawyers and other stakeholders in the sector. Regulatory mechanisms should be put in place with regards to actions that will be undertaken by AI-powered systems, to ensure that legal practitioners supervise the output given by these systems so as to avoid liability. There should be different metrics and regulations laid down by the regulatory agencies to ensure that no form of prejudice or bias is fed into the AI-powered systems, so as to prevent any form of distortion of the output given by the AI-powered system. It is also pertinent that a comprehensive review aimed at eliminating every form of bias and prejudice be made on all AI-powered projects in the Legal sector. The Data protection regulations should have strong enforcement mechanisms so as to safeguard against the compromise of sensitive data as it relates to Clients.
















The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.

See More Popular Content From

Mondaq uses cookies on this website. By using our website you agree to our use of cookies as set out in our Privacy Policy.

Learn More