Rumblings of a revolution
Artificial intelligence may transform the profession
- March 29, 2019
If you were asked to name an occupation that soon will be transformed by artificial intelligence, accounting probably isn’t the first one that comes to mind.
Nonetheless, CPAs and accounting professors say that AI — combined with other technologies such as blockchain — may radically change the profession’s procedures within the next decade.
“The accountant of today is going to be different from the accountant of tomorrow — and far, far different from the accountant of yesterday,” says J.K. Aier, an associate professor and area chair of accounting at George Mason University. “Things are changing as we’re trying to understand it. … It’s very difficult to predict what an accountant will be doing 10 years from today.”
Global accounting firms such as Deloitte, PricewaterhouseCoopers and KPMG already use machine-learning software to handle some auditing tasks that entry-level employees typically handled, such as reviewing lease contracts.
These firms also are investing hundreds of millions of dollars in proprietary artificial intelligence technologies that will be used in the 2020s, says Gary Thomson, Richmond-based regional managing partner for Dixon Hughes Goodman, the largest accounting firm in the South.
Because of these trends, a national professional association, the American Institute of Certified Public Accountants (AICPA), announced in October that it’s developing an AI-based auditing platform, Dynamic Audit Solution (DAS), which will be available to all of its 14,000-plus member firms. Funded with $50 million from AICPA members, the DAS initiative aims to level the playing field so that AI technology won’t be out of reach for small firms.
“There’s [almost] 15,000 firms in America that do audits, and a lot of them are small- and mid-sized, and we wanted to make sure that they also had the opportunity to leverage these new capabilities,” says Erik Asgeirsson, president and CEO of CPA.com, AICPA’s technology subsidiary. “It’s an exciting time for the accounting profession due to how technology is really allowing them to enhance their value proposition.”
AI technology, Asgeirsson says, will allow accountants to examine all of a client’s financial records during an audit — no matter how large the data — instead of just reviewing sample sets.
“You’re going to determine things that you weren’t able to determine in audits that weren’t leveraging these platforms,” he says. “The modern audit is going to allow the auditor to tell the client that they’ve looked at the full data set … [and] found such things as double-billing errors, payments that weren’t being done correctly … and they’ll basically be giving the client much more assurance around the integrity of their data.”
In discussing the potential impact of AI on accounting, industry experts often mention blockchain in the same breath. Developed in 2008 for the cryptocurrency bitcoin, blockchain is a system for encrypting a secure ledger of transactions that — in theory — cannot be altered.
Used as an accounting tool, blockchain could allow tamper-proof, real-time access to a company’s financial transactions at any time. Combined with the ability of advanced AI to review all records, blockchain could be a game-changer for corporate financial auditing.
“I’ve been quite nervous about [blockchain],” says Ashley Austin, an assistant professor of accounting at the University of Richmond. “It’s supposed to be this perfect ledger. And if you have a perfect ledger, why do you need an auditor to come audit it? Why do you need accounting majors to go work there and a professor to teach them, etc.?”
Thomson has heard similar concerns about blockchain “and rightfully so on the part of practitioners about: Does this in some way potentially inhibit growth in the CPA world, particularly in the auditing world?”
Nonetheless, Thomson and Austin say regulators probably will need to review blockchain and some AI technologies to ensure they can be used safely and securely for auditing. Such a delay may give the industry more time to adjust.
Fewer data-entry jobs?
A greater likelihood in the short term, experts say, is that artificial-intelligence technologies may result in bigger accounting firms hiring fewer employees for jobs such as doing data entry, reviewing leases and providing third-party verifications of a client’s financial transactions. These rote tasks can be handled by AI and optical-character recognition software.
“You probably can eliminate the data-entry people. You can eliminate probably the first line of supervision. Whereas before you may have had several accounts-payable supervisors, now you may only have one [person] running automated accounts payable,” says Douglas E. Ziegenfuss, chair and professor of accounting at Old Dominion University.
Some large accounting firms now are recruiting tech-savvy accounting majors with dual majors in information technology or business analytics. These recruits can utilize the latest AI tools and analyze the findings.
“They need to be thinking a little more like a scientist than accounting majors have in the past, coming up with hypotheses,” Austin says.
In some cases, Ziegenfuss says, CPA firms are hiring tech workers and training them in accounting. “They’re sending them back to schools that cater to nontraditional students like ODU to pick up the accounting,” he says. “They feel it’s easier to do that than to take an accountant and make an IT or business-analytics person out of them.”
Driven by this hiring pattern, many universities are adjusting their accounting curriculums to include an increased emphasis on data science and analytics.
GMU, for instance, now offers a graduate certificate in accounting analytics aimed at professionals wanting to upgrade their skills. Likewise, Virginia Tech began a partnership with KPMG last fall to offer a one-year master’s degree in accounting that includes training in data and analytics. One of nine universities nationwide working with the firm on such a program, Tech receives training from KPMG and access to its proprietary machine-learning analytics software. That software also is used in a senior-level, undergraduate accounting analytics course.
Students use KPMG’s software “to work with big data in terms of dealing with hundreds of thousands and millions of records of observations and being able to get comfortable in working with that. And then also being able to display the results so that it makes sense to decision makers,” says John J. “Jack” Maher, head of the Department of Accounting and Information Systems at Virginia Tech’s Pamplin College of Business.
Anticipating the big technological changes to come, everyone in the profession “needs to up-skill,” Asgeirsson says. “We’re all going to have to put in 100 hours of learning, 100 days of learning, over the next four or five years. ... You just can’t rest on what you’ve learned because [the accounting] business and operations are changing.”