Contracts have been the backbone of businesses since time immemorial. They form the basis of any business transaction, making it essential for them to be accurate, comprehensive, and legally binding. Contract drafting is a complex process that requires thorough legal knowledge and attention to detail. However, it is time-consuming, expensive, and can lead to costly errors. This is where Artificial Intelligence (AI) comes in to revolutionise the legal industry. In this article, we'll explore AI contract drafting, how it works, and its pros and cons.
AI contract drafting is the process of using AI technology to create, review, and manage contracts. This solution applies machine learning, natural language processing, and deep neural networks to help lawyers streamline the drafting process.
AI contract drafting software leverages data analysis capabilities to provide feedback, proofread, and improve contract language. By doing so, it reduces the risk of errors and inconsistencies, enabling lawyers to spend more time on high-level tasks that require a personal touch.
AI contract drafting involves developing contract templates that can be customised to suit a particular transaction. Once the template is created, the AI software uses natural language processing to understand the content and context of the contract. It then scans multiple databases to retrieve relevant clauses, case law, and legal language, which it uses to generate a comprehensive contract.
AI software is also capable of analysing the language used in existing contracts to identify patterns and provide recommendations. It can flag discrepancies, loopholes, and other issues that may not be noticeable to human lawyers, enabling them to make adjustments and avoid legal risks.
The use of Artificial Intelligence (AI) in contract drafting has become increasingly popular, offering several advantages but also presenting some challenges. Here's a look at some of the pros and cons.
Efficiency—AI can generate contracts faster than a human, enabling companies to create and execute contracts more quickly. This is especially useful for standard contracts that don't require significant customization.
Cost Savings—Automation usually reduces the cost of contract drafting, particularly for businesses that need to create large volumes of similar contracts.
Accurate Language—AI tools can reduce the likelihood of human errors, such as typographical errors or inconsistencies in language, that may compromise the contract.
Automated Analysis—AI contract drafting tools can analyse contracts to identify potential risks or areas that may require negotiation, aiding in better decision-making.
Easier Updating and Adaptation—AI contract systems can easily update contracts to reflect changes in law or company policy, helping to maintain legal compliance.
Accessibility—With cloud-based systems, contracts can be drafted, edited, and shared from anywhere, facilitating better collaboration among stakeholders.
Data-Driven Insights—AI algorithms can analyse large sets of contracts to provide insights into market trends, negotiation tactics, and other valuable information.
Lack of Nuance—AI-generated contracts may lack the nuance and understanding of human emotion and strategic considerations, which are sometimes crucial in negotiations.
Limited Creativity—AI systems are not good at thinking outside the box, so they might not be able to draft contracts for novel or complicated situations without human intervention.
Legal Risks—Poorly calibrated AI could generate contracts that are non-compliant with local laws or that contain unintended loopholes, creating potential legal risks.
Depersonalisation—The use of AI in contract drafting could potentially harm relationships with clients or partners who may value the personal touch in contract negotiations.
Training Complexity—AI tools often require an initial training and playbook setup phase to adapt to a company's specific needs, which could be resource-intensive. “It's going to take a lot of training for you to be able to be confident that the model will produce the drafting that you really wanted to have produced in any given scenario,” says Tim Pullan, CEO of contract acceleration platform ThoughtRiver which is trained by legal experts on more than 1500 Lexible documents.
Cost of Implementation—While they may reduce costs in the long run, the initial investment in AI systems can be high, especially for high-quality systems.
Ethical Considerations—There could be concerns related to job displacement due to automation, or questions about how the AI was trained (e.g., was it trained on contracts that may perpetuate existing biases?)
Dependence on Data—Most of the commercially available AI algorithms are only as good as the data they are trained on. Incomplete or biased data sets can result in flawed contracts.
Ongoing Maintenance—Like any other software, AI tools require regular training, updates and maintenance, which can incur additional costs and downtime
The suitability of AI for contract drafting will vary depending on the complexity of the contracts, the specific needs of the company, and other contextual factors. Many organisations adopt a hybrid approach, using AI for more routine contracts while relying on human expertise for more complex or sensitive negotiations.
The accuracy of AI contract drafting can vary widely depending on several factors, and the main one is the quality of AI algorithms. Well-designed algorithms, especially those trained on large, high-quality data sets, are generally more accurate than commercial-grade, general-use AI chatbots like ChatGPT.
Also, the adage “garbage in, garbage out” still applies. If the AI has been trained on poor quality data or outdated contracts, it will likely produce less accurate drafts. AI systems that have been well-configured and trained on a company's specific templates and guidelines will typically produce more accurate results.
While AI can reduce human error in terms of typos and omissions, they can also introduce their own errors, particularly if the algorithm misinterprets a legal clause or is not updated in line with legal changes.
In general, for standard contracts that don't require much customisation, AI can be quite accurate. However, for complex, nuanced, or highly specialised contracts, human expertise is often needed to ensure accuracy. Some areas of law are more complex and subject to more frequent changes and interpretations than others, which can pose a challenge for maintaining high levels of accuracy.
But regardless of the algorithm, AI contracts often benefit from human review to catch nuances, consider strategic implications, and ensure compliance with current laws, which can differ by jurisdiction and change over time.
“Contract drafts is not casual creative content. Its core requirement is to be precise,” says Tim Pullan, CEO of ThoughtRiver, “And for anyone looking to rely on generative AI to either redraft as an amended document or create a document from scratch, the level of work that they will need to put into that output to make sure that it is exactly what they want is going to be necessarily higher because of the that precision requirement that you have with contracts.”
Given these factors, many organisations adopt a hybrid approach, leveraging AI for efficiency and scalability while relying on human expertise for nuanced review and to ensure legal compliance.
AI frameworks offer an optimistic outlook for the future of contract drafting, bringing advantages like enhanced speed, fewer mistakes, and broader accessibility to the table. Nonetheless, for the time being, it's crucial to find the right equilibrium between adopting these advanced tools and retaining human discernment.
The use of a hybrid approach that combines both human expertise and AI capabilities can offer the best of both worlds in contract drafting. Here are some best practices for such an approach:
Scope Identification—Determine what types of contracts or sections thereof are suitable for AI automation and which require human oversight or drafting.
Legal Compliance—Ensure that the use of AI in contract drafting complies with local and international laws and ethical standards.
Stakeholder Buy-In—Get buy-in from legal, business, and IT teams to ensure a cohesive approach.
Automate Routine Tasks—Use AI for drafting standardised clauses and sections that do not require customisation or nuance.
Human Oversight—For critical clauses or complex contracts, have legal experts draft or review these sections.
“Obviously in theory there's a very large time saving there because a lawyer might take two or more hours to amend a contract. But the problem there is your ability to control the output and to be confident that what it produces is what you want. The amount of time it takes you to check the output could eliminate all the time-saving gains and even worse,” says Tim Pullan.
“For example, a U.S. law firm partner who'd been trialling a generative drafting tool, took a 12-page contract and said to the generative tool, hey, can you redraft this. What came back was a 20-page document, so it almost doubled the size of the document. And in reality, it took him longer to produce what he wanted because he then had to check all this additional content, some of which he didn't want.
"Because generative AI is there to produce content, right? Given a chance, it will produce more content, which means more amendments, and every single one of those amendments reflects a piece of scope in that negotiation which is only going to delay it."
Tim Pullan, CEO of ThoughtRiver
Quality Checks—Implement a multi-tier review process where both AI and human experts validate the contract for completeness, accuracy, and compliance.
Transparency—Be transparent with stakeholders, including clients, about the extent to which AI is used in contract drafting.
Job Roles—Clarify how the role of human experts will evolve with the adoption of AI and offer training for the transition.
Bias Mitigation—Regularly review and update the AI training data and algorithms to minimise biases and ethical risks.
Version Control—Maintain a robust version control system for both AI-generated and human-reviewed contract drafts.
Audit Trails—Keep a detailed record of changes, approvals, and reviews for compliance and future reference.
Metadata Storage—Store relevant metadata like the author, reviewer, and date of last modification for each contract.
By following these best practices, organisations can optimise the drafting process, achieving greater efficiency and reliability without sacrificing the quality and integrity of their contracts. As legal frameworks continue to change, fusing AI capabilities with human skills will be key to successfully manoeuvring through the intricate domain of contract creation.
While AI in contract drafting is generally heralded for its efficiency and speed, there are scenarios where it may not lead to improved efficiency. For example, AI-generated templates may require extensive human review and editing, thereby reducing efficiency. If clients or stakeholders don't trust the AI system for any reason, additional time may be needed to manually review and confirm contracts, reducing the system's efficiency gains.
Most importantly, if the AI does not have robust quality control mechanisms, its errors may go unnoticed until later, requiring time-consuming corrections.
“If you really want to speed up the time to signature, there are factors which far outweigh any saving that you're ever going to get from speeding up a lawyer. How many issues do I decide to negotiate? Do I decide to fight on twenty, thirty, ten points? Every single one of those points for your counterparty means getting involved in the negotiation,” says Tim Pullan.
“In most of the cases, there is a need to go to internal stakeholders and get their approval decision on that point. If you're looking at the efficiency of a transaction, how quickly you produce content is not going to move the needle. What really moves the needle is being more efficient with the scope of what you decide to negotiate on any particular contract.
The air gaps arise when people are waiting for other people to do things. So, if you've got thirty issues, all of which need to be discussed in collaboration with internal stakeholders, this can give rise to delay when the lawyer's waiting for an internal business stakeholder. That’s where the inefficiency in contract negotiation happens. And if you decide to use generative AI to redraft the contract, you're likely running into the significant risk that you're creating more issues.”
Given AI's benefits in law and its ever-growing advancements, the future of AI contract drafting looks bright, despite some growing pains. As we move towards a more digital-driven legal landscape, the need for smart contract technology that utilises AI-powered contract drafting solutions becomes both necessary and inevitable.
There is a merging need to complement human expertise and technology-powered solutions that can leverage data-driven insights. As such, legal firms must embrace and implement AI in their business models to stay competitive and relevant.
In summary, AI serves as a complementary tool in the realm of contract drafting, enhancing both the work of legal experts and the experience of non-legal professionals. For seasoned lawyers, AI automates routine tasks, allowing them to focus on complex legal intricacies and strategic advising.
Meanwhile, for those without a legal background, AI simplifies the drafting process by providing standard templates and flagging potential areas of concern, making contract creation more accessible and less daunting. By streamlining repetitive tasks and providing intelligent suggestions, AI technologies elevate the efficiency, accuracy, and inclusivity of the contract drafting process for all involved.
With its sophisticated data analysis capabilities and continuous learning and improvement, AI contract drafting is fast becoming essential in drafting simple individual and mass contracts. The future of AI contract drafting in law looks promising, and the rewards of perfecting its algorithms are significant.
"People should absolutely try to use generative AI in their work, but they should be prepared to experiment a bit to find out what works for them. My advice generally would be staying in control. This way, you're more likely to reach productivity benefits."
Tim Pullan, CEO of ThoughtRiver
ThoughtRiver is an AI-enabled contract acceleration platform that empowers your team to review and manage contracts in one secure portal. Trained by human legal experts on over 1,500 Lexible legal questions and billions of contract pages, ThoughtRiver is the industry's most accurate legal AI contract review platform.