How—and why—EY uses AI in the audit process

We could keep telling you that AI is going to fundamentally change financial professions—or you could take it from someone right at the source, like Richard Jackson, an assurance partner at Ernst & Young.

EY has already started using AI tools in the audit process—and Jackson’s insights might help you or your organization as you consider similar investments.

Jackson started his career at EY in the 1990s in the UK, but he said the lure of “this mythical place called Silicon Valley” brought him to California, where he’s worked with technology clients ever since.

“I’m strongly of the belief that we’ll see more change in the next five to 10 years in this profession than we’ve seen in the last 100,” Jackson told CFO Brew. “That’s hugely energizing for any CFO that’s trying to lead, encourage, and motivate their own organization.”

He notes that while AI exploded into the public consciousness within the last year or so, EY had already been in internal talks for years to determine the best way to deploy the emerging tech. Since those early conversations, Jackson said the emphasis has always been on ways to make work easier for people, rather than replacing their jobs.

“It’s about using that technology and automation to help create more time and space for the auditor to apply their judgment, experience, knowledge, and expertise,” he said.

In front of the auditor. What does that look like in practice? EY uses an in-house audit technology application, called EY Canvas, which was developed with Microsoft. That tool allows auditors to develop a unified plan across different teams through a global audit platform, while allowing for real-time monitoring and quick changes.

”In a nutshell, it basically brings the intelligence and the insights around your client, your industry, [and] your peers to the hands of the auditor when they’re thinking about the scope and the risk assessment involving the audit,” Jackson explained. It captures a wide swath of public information, compiling things like short sell reports, and puts that information in front of the auditor, according to Jackson.

While auditors could hypothetically sift through the same data, the depth of that information and the time needed to properly inspect it often aren’t feasible for a given audit. Without the help of AI, potential risks can fall through the cracks. But with this increased speed, EY can now better detect fraud in more sophisticated ways than an individual person could.

“If you think about the impacts of fraud over the last few years, it’s quite often been associated with red flags that, along the way, perhaps have been missed by investors, by companies, sometimes even by the auditor,” Jackson maintained.

EY’s tech is meant to make those red flags easier for auditors to spot, he explained. Rather than looking at a small or random sample, the technology helps auditors better observe total populations, he said, allowing them to locate if a particular risk falls within one business unit or product category, for instance.

New approach. With this advance, EY has also had to change the way it handles data, Jackson noted. “We moved [from]…the old school approach of ‘pick a sample of 25 and bounce it back to the supporting documents,’ to a world where we can now capture total populations of data from our clients’ ERP, from their finance systems, and operations,” he explained.

Now, he continued, tens or hundreds of millions of records can be analyzed to identify what constitutes a normal transaction. Before, you’d have to rely on some amount of luck in finding an accurate random sample to give you a sense of what “normal” might be. Crucially, the bigger sample set also makes it easier to spot transactions that fall out of line.

That helps point auditors to transactions they “need to take a close look at because that’s where the risks might apply,” Jackson said. “The technology allows you to look across the totality of a transaction population, and actually, instead of just hoping that you happen to choose the right sample, is now actually able to surface all the areas where the auditor can follow up on.”

Learning from others. While EY’s specific methods are tailor-mode for the firm’s audit needs, Jackson encourages “all CFOs to have these conversations with their auditors to understand where their auditors are in this journey.”

“Those are the same opportunities, challenges, and conversations that CFOs have for their own organizations,” he said.