How to Be Ready for Real Artificial Intelligence

What to know before investing in new technology.

By Hitendra Patil and Eli Fathi

To use AI or not?

That is the question for accounting and audit firms looking toward the future. There is a lot of fear, uncertainty, and doubt regarding the availability, adoption and use of artificial intelligence-based systems. There is even some misleading information within the broader community of accounting and audit tools.

MORE ON ENTREPRENEURIAL STRATEGY: How Accountants Can Ride the Artificial Intelligence Wave | The Client of the Future | The Rise of the Robot Accountant: Opportunity or Threat?
GoProCPA.comExclusively for PRO Members. Log in here or upgrade to PRO today.

A wide range of claims and news can be found online, from AI “not being ready” for accounting, to questioning the benefits of artificial intelligence. Vendors stating their solutions are AI-enabled may be adding to the confusion.
When and which?

Accounting firms must decide when to invest in AI, and which tool(s) could provide the best return on investment. Business leaders can better navigate their technology adoption path by understanding how AI fits into their technology infrastructure and audit processes. They also need to differentiate between solutions that include real AI and machine learning, and those that don’t.

The case for artificial intelligence

Before adopting AI, firms need to understand how the technology will improve their business and boost value to clients. Let’s take an example of audit services. Examining the firm along the following criteria will help answer how an AI-enabled solution(s) will help:

  • How many audits does the firm perform per year?
  • What is the average audit size (in the number of transactions)?
  • What is the cost of performing those audits?
  • What level of risk does the firm take on with those audits?

Beware. It’s easy to conclude that AI is best suited to large firms that perform audits at a greater size, cost, and risk.

But the reality is that firms of all sizes should proactively consider using AI.

The greatest values of AI come with the ability to digest and understand enormous amounts of information much faster than any traditional technology and that all transactions are subject to an analysis that goes beyond rules-based or statistical methods. AI reduces the time and costs associated with data analysis and provides new ways of reporting on completeness, thoroughness and the risk of material misstatements.

AI-based solutions eliminate the need for sampling as they can provide 100 percent transaction coverage. AI also identifies risks based on learning and adapting to the dataset, far exceeding the results of traditional rules and statistics. To evaluate the cost performance assessment of different tools, firms must compare the engagement costs and the degree of risk associated with providing opinions to clients with AI, and without.

AI helps firms of all sizes

Should only large accounting firms use AI?

The answer is no. The decision depends on the firm’s engagement costs, core competitive advantages, and how AI improves them moving forward.

In the past, the challenge for smaller firms was the cost and learning curve required of leading-edge technologies. Today, we are seeing the “democratization” (uberization) of AI that enables all firms to leverage the technology with cost-effective solutions and human-centric user experiences.

As with any new technology deployment, firms should consider how people are trained and supported by the AI solution provider. There are costs associated with this, and onboarding time needs to be allocated for both users and business leaders. It’s important to know if the AI solution provider has a robust adoption program and support services that fit the organization’s needs.

Critical change areas for firms to consider are ingesting data from client systems, mapping product workflows to audit processes, and reporting to the client. The time spent on rolling out new technology is also important, as that can take resources away from working on the firm’s business. The software must be really user-friendly, and the solution provider must be willing to transfer knowledge across the firm, training auditors and partners “how to fish” so they can cut the umbilical cord of dependency. It is a classic business case. The new value provided by the solution should be worth the investment and the solution provider should have a clear plan to get you to the benefits fast.

After bringing AI into her small firm, Samantha Bowling – CPA, CGMA, and partner at Garbelman Winslow – said, “It gives me a comfort level that I’m looking in the right place.” Bowling also uses AI to decide whether to take on a new client or not by examining their general ledger using the technology and charges clients more for its use.

Researching the AI solution provider is critical

Investing in new technologies is a journey and firms must pack the right items to ensure success. A major source of information, perhaps the most important, is the provider of the AI solution. As AI for accounting is a comparatively new technology and causing some confusion in the landscape, firms need to consider items beyond cost, training, and support. They must also determine whether artificial intelligence exists within the tool and how it will improve the way they perform audits over current methods.

How do firms know they are getting real AI?

Anyone can say that they have AI-based solutions or that AI is “built into” their tool but how does someone not familiar with the technology know?

There are several areas to examine. For example:

  • Avoid the “black box” syndrome, in which vendors claim that they have AI, but it’s not evident in the behavior of the tool. This doesn’t mean that there’s some explicit indication of “AI” in the user interface (in fact, the tool should err on the side of user-friendliness rather than technical complexity) but vendors should be able to explain how they utilize AI in their tool and to show you how it enhances results.
  • As AI is introducing new ways of determining risk and providing new and better value to clients, the vendor should be able to explain their algorithms and how they relate to risk reporting.
  • They should also be transparent as to how these algorithms go beyond traditional rules-based testing and statistical methods. If all they offer is Benford’s Law and weekend post checks, the value is most likely not there.
  • It’s important to know whether the vendor has actual AI experts on staff. Roles to look for, including on their careers page, are data scientists and machine learning experts.
  • Equally important is to determine how long has the product been used by audit firms in the field.

The bottom line is that firms should obtain the information necessary to have a high degree of confidence that AI actually exists in the platform and that it is used to create new and differentiating value.

AI-based solutions are designed to make firms more competitive and offer a level of risk assurance to their clients that were never possible before. It’s crucial for business leaders to
select the solution that is offering AI right here and right now, seeing through the noise to best
position their firm for success.

Co-author Eli Fathi is CEO at MindBridge Ai, developer of the world’s first auditing tool based upon artificial intelligence and machine learning technologies to uncover errors in financial data.