Reflections on IACCM – Part 2: Artificial Intelligence
In my previous blog, I discussed the current state of adoption of CLM systems into enterprises across India. In this installment, I’ll answer the questions I received at the IACCM conference in regards to artificial intelligence (AI) and its impact on legal departments.
Will AI take away jobs in the legal department?
Technology has always been feared because it replaces humans with automation. History is replete with such instances – horses were replaced by automobiles, low-end workers in assembly lines in manufacturing industries have been long replaced by robots, recently robots have started replacing humans in most manufacturing processes in industrial plants in North America and the EMEA region. However, if technology takes away low-end jobs then it also ends up creating many more high-end jobs. Specifically – speaking about legal – simple and routine automated jobs would likely get replaced with the increased adoption of contract management technology. But the bright side is that qualified legal resources would have a lot more time and bandwidth available to spend time on high-impact and high-worth activities in the legal department. Those who embrace technology would leverage it to move up in their career, while others will become victims if they allow this opportunity to pass by!
How is AI disrupting the way corporate legal works?
AI is yet to disrupt the existing methods of corporate legal in India, but enlightened legal executives are asking the right questions to identify how they can leverage AI/ML (machine learning) to improve performance. There is a lot to consider with your existing contract management technology before considering adding AI/ML.
A frequent question (even respected CIOs have asked), “How can a prospective enterprise business leverage AI in a contract management solution?” My answer includes: “The first requirement is to get your foundation strong and establish standards for your contract language and data set”. Those who ignore this fundamental truth have not seen the cost/benefit in the process. For machine learning algorithms to start aiding your decision making, one needs to first design and then train on a model to start yielding the desired level of confidence so you can trust the outcome. The fundamental rule in the science of probability (part of statistics) is that the higher the confidence level on a scale of 1 to 100, the better the predictability of the algorithm.
Contrary to popular sentiment, there is no “Plug-n-Play”. To enable AI-infused decision making or automation, one has to first select the appropriate algorithm that is relevant to the task and then train the algorithm on past business data (more the better – rule of thumb is that you should have at least six months of data to ensure an accurate output). Then, fine-tune the algorithm. Then, it is up to the business executive to decide what level of confidence works for them. The usual standards are between 85% and 90%. Some businesses use a 60% confidence level because it still saves a lot of time for their legal teams. Therefore, before getting into the strategic arena of deciding and investing in AI to drive legal processes and decision making, one should have a well-defined business use case for leveraging AI in decision making.
In part 3 of this blog series, I’ll share some interesting CLM use cases I have encountered during my travels.
Are you ready for the Future of Contracts? Learn key trends from the experts.
Download our whitepaper today.