Elon Musk: “Mark my words — A.I. is far more dangerous than nukes”
While Mr. Musk could be right about the above, his concerns are mostly about Artificial “General” Intelligence that trains computers to mimic the problem-solving ability of human beings in everyday scenarios. However, Artificial “Narrow” Intelligence is focused on algorithms that specifically solve a certain problem. An example would be using AI to predict and mitigate payment frauds. These applications of AI have actually helped in creating jobs.
To validate the above point, a D&B survey at the AI World Conference and Expo found that 40% of organizations are adding more jobs as a result of deploying AI in their organizations while 34% reported that job demand has remained the same.
For the purpose of simplicity, whenever we use AI in this blog, we are referring to Artificial Narrow Intelligence. We will discuss how AI is working its way into finance departments and why it should be seen as a friend and not a foe.
AI, machine learning, and RPA are often used interchangeably by technology providers, and it is difficult to grasp their true meaning and how they are distinct from one another. The following definitions aim to clear the air around these technologies and their capabilities.
“Artificial Intelligence is the science of making machines do things that would require intelligence if done by men.” Marvin Minsky
Artificial Intelligence or AI is the ability of computer systems to learn, reason, think, and perform tasks requiring complex decision making. A simple example of AI is how Google uses real-time data to find you the shortest route to your destination.
Generally, AI is used as a parent term, and Robot Processing Automation and Machine Learning cover subsets of applications of AI.
Robot Processing Automation or RPA is the use of software “robots” mimicking human actions to perform a well-defined business process.
RPA is best suited for performing tasks that are repetitive in nature and executed by following a fixed set of rules. RPA, for example, could be used to read data from paper-based forms and then insert the text into customer applications to trigger workflows for further processing.
Machine Learning is an application of AI, capable of identifying patterns from a large set of data with the help of algorithms. It is self-learning in nature and becomes “smarter” over time.
An example of machine learning would be the algorithm that Amazon uses to show you hyper-personalized product recommendations that you could purchase the next time you visit amazon.com.
It could be deployed to predict future outcomes and identify trends.
All business processes rely on human activity – primarily for abstract and complex decision making, but unfortunately also for a lot of repetitive and clerical tasks – such as matching transactions or clearing invoices.
Artificial Intelligence is already solving this problem for dozens of industries including CPG, healthcare, media, and insurance. The ultimate goal of AI is to improve the quality of human life.
AI’s purpose is not to replace humans, but for humans to focus on more critical problems or problems that they never had the time to focus on. For example, in a very deductions-prone industry like the CPG, imagine how much margin erosion AI could prevent if analysts are given recommendations on valid and invalid deductions.
DSO is one of the most important metrics that senior management focuses on, especially in the current macroeconomic scenario of rising interest rates. However, the leadership team only has a few levers to pull to impact DSO. With AI, the management team is able to predict the impact of a pre-payment discount policy on DSO six months down the line. AI, in this case, is also aiding decision making by helping executive teams forecast the impact of a change in the credit policy.
Today, process managers are primarily focused on ensuring that their teams are operating at maximum productivity. They keep close tabs on the performance of individual analysts and constantly reallocate time-sensitive work in the event of resource shortages or absence.
In the future, with 90% of the transactional activities running on auto-pilot, analyst productivity will no longer be a concern for process owners. AI could enable them to closely analyze processes, identify signs of distress, and administer course correction and long-term improvement.
A collections manager could, therefore, be less concerned about CEI and more about whether their teams have built the right rapport with critical accounts and help in dispute resolution.
Proactive collections and dispute resolution are primary responsibilities for any A/R or deductions analysts. But in the current scheme of things, analysts spend the majority of their time in low-value work, including:
All these tasks are resource-intensive and do not create significant value.
AI is already automating a lot of these low-value activities and enabling analysts to focus on high-value tasks, such as building better customer experience to resolve disputes and reduce delinquency, recovering invalid deductions, and controlling credit risk, rather than just chasing transactions and spreadsheets.
This blog just touches the tip of the iceberg as far as showing the possibilities of AI in O2C is concerned. While there might be some fears about AI leading to job losses, I hope that I have dispelled some myths and made a case for why AI makes jobs better and unlocks new possibilities.
Do let me know in comments if you have experience with creative application of AI in the finance function.
Looking for more information on how AI and RPA could transform finance? Download a complimentary eBook here.
11 Proven Strategies to Reduce DSO
10 Pressures On Finance Shared Services Today And How To Survive… Indeed Flourish