With an explosion of the popularity of ChatGPT, artificial intelligence (AI) has assumed a place at the center of consciousness of many business executives. These executives have begun to think about how the application of AI can impact their business both positively and negatively. Many pundits and media personalities have asked serious – and somewhat unserious – questions about the role of AI in society. AI has recently passed exams for medical school, law school, and business school. Notably, AI tried to convince one reporter to end a personal relationship.

While there has been a lot of recent excitement about AI, the thoughts of how to deploy it in support of business have largely trended toward the theoretical. Certainly, a high school student can use it to “write” an essay for a class, but the direct business application of AI to an accounting function is less clear. Larger societal questions aside, what is the business case for the use of AI in the modern accounting function?

AI Is Not New, but It Has Changed

Keep in mind that AI is not a new concept. AI has been in use for years, even decades. Several of the major enterprise resource planning systems have long incorporated AI. Most of the major robotic process automation (RPA) programs have AI either built into the base program or can easily add it on for an upcharge.

What is new is the way in which humans can interact with AI. Services like ChatGPT make it easy for an average accountant to marshal the full power of AI to serve their business needs without the need to earn an advanced degree in computer science. This is another example of a theme I have expressed for several years: These “new” technologies are not actually new; what has changed is the ease with which the average person can interact with these technologies and invoke their full power.

The Business Case for Leveraging AI

So, if the “average” accountant can easily leverage AI for use in day-to-day operations, what is the business case for doing so? To answer that, let us ask the AI.

(Note: All italicized words in this article are from ChatGPT)

One of the primary benefits of AI in accounting is its ability to automate repetitive tasks, such as data entry, invoice processing, and financial reporting. In addition, AI can reduce errors and increase accuracy by eliminating manual data entry and automating the reconciliation of financial statements.

AI-powered fraud detection tools can analyze large volumes of financial data and identify anomalies that may indicate fraudulent activity.

By analyzing historical data and trends, AI algorithms can predict future outcomes and identify potential risks and opportunities.

This sounds wonderful. Are there any areas for concern?

One of the primary concerns is the risk of job displacement, as AI-powered tools can automate many tasks that were previously performed by humans.

Another concern is the risk of data breaches and other security threats.

In addition, there is a risk that AI algorithms may produce biased or inaccurate results if they are trained on incomplete or biased data sets. It is important for businesses to ensure that their AI systems are properly calibrated and tested to minimize the risk of biased or inaccurate results.

Based on AI’s assessment of itself, it seems that many large risks remain. It would obviously be foolish for an accountant to feed a set of unaudited financial statements into ChatGPT (a public, unsecured system) and ask it to look for issues. Rather, care must first be taken to secure an AI program and then develop it to suit a business’ needs. This will require the inputs of accountants, IT professionals, security professionals, and operational professionals to ensure that the resulting product is secure and useful.

Next, significant effort must be deployed to train and test the AI. All these efforts rely upon the underlying data being complete, accurate, and free of bias. Data quality has been an issue for data analysts for as long as data has been analyzed, and the AI will only be able to express outcomes of quality in line with the quality of the inputs.

AI and Productivity Gains

Here is the way to perceive AI: It is simply the next efficiency tool that accountants can utilize to become more productive. Just as spreadsheets increased productivity and the 10-key before that, AI will be the next tool that accountants will use to “do more with less.” This is not new and absolutely should not be “scary” to the modern accountant.

Productivity gains are the true drivers of economic growth, and the modern accounting function will be in the center of the AI evolution. Likely, we will continue to see AI being successfully deployed within other programs, as it is presently deployed within RPA software and data analysis software. Keep this adage in mind:

My caveat: This modern accountant who knows how to deploy AI should be you.