Striking the Balance: Leveraging AI for Effective Contract Reviews Without Overcomplicating

Striking the Balance: Leveraging AI for Effective Contract Reviews Without Overcomplicating

Artificial intelligence (AI) is increasingly being leveraged to streamline various tasks and its application in contract reviews is a topic of much discussion.

Recently, Nir Golan's comments on my experience testing Google's Gemma small language model sparked some interesting thoughts. He questioned whether the plethora of suggestions touted by current AI review tools is genuinely beneficial. This concern is a valid one, particularly in light of experiences where the drive for optimisation led to unnecessary complexity.

At the heart of integrating AI into contract review processes is the quest to make our lives easier. However, the risk of getting bogged down in an overwhelming flood of suggestions cannot be ignored. The scenario is reminiscent (at least to me) of less experienced developers designing solutions for non-existent problems. I recall instances where developers prepared an internal app for an audience of 100 million users, despite the realistic expectation of not exceeding 100 users. The effort to scale for a million times the realistic user base introduced needless complexity. This analogy highlights the ease with which over-preparation can complicate matters, offering a parallel to the potential pitfalls of AI in contract reviews.

The challenge, then, lies in setting the AI's review parameters judiciously. The focus could range from specific word usage and grammar to both, given that these factors significantly influence the meaning of contract clauses. It's vital, however, to avoid getting entangled in minor details that contribute little to the contract's overall clarity and fairness.

To navigate this, setting AI review parameters might involve directives to:

  • Identify and underline clauses that disproportionately favour one party over the other.
  • Review common contract elements, such as indemnity, limitation of liability, and termination clauses, ensuring they adhere to standard practices and serve the client's best interests.
  • Assess the contract's clarity and conciseness, advocating for straightforward language over complex legal jargon.

Consider an AI suggestion that transforms a simple statement like:

This agreement requires both parties to work together effectively to achieve their common goals.


This agreement mandates the utilisation of robust paradigms to ensure enhanced synergy among the parties involved.

While the latter might sound more sophisticated, it adds unnecessary complexity without improving the contract's substance. This tendency to "upgrade" language, often seen when giving minimal prompts to AI like "Make this sound better," can lead to outputs that are more verbose than valuable.

The overarching goal should be to harness AI in a way that enhances efficiency without sacrificing substance. By carefully crafting prompts and understanding the appropriate depth of review, we can employ AI as a tool that genuinely lightens our workload. This approach ensures that AI aids in streamlining the contract review process, providing meaningful suggestions that enhance clarity and fairness, rather than inundating us with superficial changes.

In conclusion, the integration of AI into contract reviews presents an opportunity to revolutionise how we approach legal documents. However, it requires a balanced approach that values substance over style. By setting clear parameters and focusing on meaningful improvements, we can leverage AI to make contract reviews more efficient and effective. This not only simplifies the process but also ensures that the contracts are fair, clear, and in the best interest of all parties involved. As we continue to explore the potential of AI in various domains, maintaining a focus on practicality and relevance will be key to unlocking its true value.