Legal Technology and Contract Drafting: from word-processing programs to Artificial Intelligence

by Matilde Serena

According to the concept of the “contract lifecycle”, transactional legal professionals engage in four stages: drafting, reviewing, managing and analysing. During the drafting stage, contracts are written in their initial form. Reviewing, then, involves the identification of legal and business terms to improve the first draft of the contract. The notion of managing relates to storing and indexing existing contracts. Analysing, finally, entails measuring the contracts’ market performance and the efficiency of their provisions. Contract drafting and managing is especially costly and labour-intensive for legal professionals. Drafting and managing technologies have been introduced since 1970; innovations in the review and analysis stages are more recent and relate to the use of machine learning.

This short essay explores the origins and early developments in the contract-drafting field, and modern-day contract-drafting programs. In particular, the focus lies on the role of AI (Artificial Intelligence). Finally, points of criticism for contract drafting AI will be presented.

Contract drafting software emerged in the 1970s and 1980s in the form of word-processing programs such as WordPerfect. The latter consisted of an office suite supporting everyday work needs, such as creating documents, presentations and spreadsheets. In the 1990s, more complex systems emerged allowing the creation of a document based on the users’ answers to a series of questions programmed according to a “logic tree”. The main issue relating to these systems was that the final document was hard-coded, meaning that the user could not make any changes to it. The only possibility was to redo the whole questionnaire again. Therefore, the use of such document assembly software was limited to standardized contracts in stable practice groups. Since the 2000s, contracting drafting software have become more sophisticated allowing users to create coded contracts by uploading pre-existing ones. These are used as a blueprint to draft similar documents. The coding of documents is, generally, done automatically through AI.

Moving on to recent developments, AI is deemed to be a tool capable of closing gaps in contract drafting and document assembly software.AI can accelerate contract drafting in many ways. Firstly, it can provide standardized contract templates and screen existing contracts more effectively than humans. Secondly, AI increases clarity and protection from disputes avoiding incurring unnecessary expenses for dispute-resolution. Thirdly, AI increases the volume of contracts negotiated and executed and offers in-depth information to assist legal professionals throughout the contract’s life cycle.

In practice, the AI software can learn from past or similar contracts by scanning them and identifying important terms and, possibly, changes. A new contract is, then, drafted within seconds and can be reviewed by the supervising legal professional. The latter can modify the template and re-submit it to facilitate the learning process of the AI software.

During contract negotiations, AI can speed up the process by identifying, organizing and analysing pertinent data to propose possible solutions, and beneficial modifications and show statistical projections of losses and gains. Moreover, AI can spot anomalies and missing clauses and estimate the duration of negotiation. Effective negotiations considerably reduce the risk of contract disputes.

After having examined the innovation and the advantages of implementing AI contract drafting software, it is worth mentioning points of criticism moved by legal professionals in this regard. Firstly, some practitioners argue that contract drafting programs limit meaningful customisation due to the fact that the underlying algorithms use a limited universe of questions and answers. This is not necessarily the case for machine learning programs as they can teach themselves to identify new examples to match the users’ preferences and improve their algorithms.

Secondly, relying on computer programs paves the way for ethical issues for legal practitioners and clients e.g. the inflexibility of these programs. To keep up with changes in the law, legal professionals will have to periodically re-code form documents and feed the software with new laws or contractual clauses.

Thirdly, legal practitioners are reluctant to transition to technology that may potentially replace their jobs. In fact, it is partially true: the use of contract drafting software could reduce the need for new associates. Nevertheless, especially machine learning programs must be trained by legal professionals to perform efficiently and are rather intuitive to use. Therefore, it is very likely that a new generation of tech-savvy lawyers will still play a fundamental role in the contract lifecycle.

To conclude, having examined the evolution of technological contract drafting tools, it emerges that using AI tools presents considerable advantages in terms of costs and workload management. Thus, there is still room for improvement, especially as to ethical issues. Nevertheless, scepticism by legal practitioners is often motivated by an unreasonable fear of being replaced by AI. This is not a realistic scenario, as AI and legal professionals shall work side-to-side to provide efficient contract drafting.

Sources:

K I Guthrie, ‘Document Assembly Software Systems’, [1995] 9 PROBATE & PROPERTY 26, 27–28.

S LaMecca, ‘Law Office Automation: A View into the Future - Automated Speech Recognition coupled with sophisticated document assembly systems will take over the document preparation function’ [1996], 45 R.I.B.J. 5, 27.

D Love, ‘What the Heck is Machine Learning?’, (BUS. INSIDER, 3 May 2014) <http://www.businessinsider.com/machine-learning-2014-5> accessed 31 January 2023.

A LeVeque, ‘Lawyers: Learn to Work with AI or Risk Termination’, (VEGASINC, 26 September 2016) <http://vegasinc.com/news/ 2016/sep/26/lawyers-learn-to-work-with-ai-or-risk-termination/> accessed 31 January 2023.

K T Betts and K R Jaep, ‘The dawn of fully automated contract drafting: machine learning breathes new life into decades-old promise’ [2017] Duke Law and Technology Review, 15, 216-233.

R Vanni, ‘How AI Accelerates the Legal Contract Drafting Process’ (Kirasystems.com, 27 May 2020) <https://kirasystems.com/learn/how-ai-accelerates-the-legal-contract-drafting-process/> accessed 31 January 2023.