Please ensure Javascript is enabled for purposes of website accessibility

Automation to Artificial Intelligence: New Frontiers for Auditors

News flash: Artificial intelligence (AI) and other cognitive technologies are eliminating jobs left and right. Bloomberg reports as many as 5 million jobs by 2020. Oh, calamity. Should auditors be worried? The short answer is no…don’t lose sleep over it.


First off, cognitive technologies (even super cool advanced ones) are best with structured tasks and finding patterns. You know, auditing grunt work. I assume most sane people would prefer to not be subjected to the mundane, mind-numbing work anyway and say, “Go ahead, automate the cross-footing and put me out of my misery!”

Computers, like IBM Watson, empower and augment auditors’ job, but are far from replacing us. For example, it’s going to be a long time before computers can manage to exercise professional skepticism — an auditor's acclaimed gut feeling that something doesn’t look right. That’s nearly impossible for a computer.

With the high employee churn in public accounting, it makes sense to me that firms would want to figure out how lighten the load of those who stick around until partner (but maybe start brushing up on some computer science theory). Plus, as a bonus, machines don’t need to eat or sleep. The potential for scalable productivity is mind blowing! 

Tom Davenport, cognitive technology thought leader and distinguished professor, agrees that AI doesn’t mean it’s all over for the auditor. He tackled the topic of AI in February in a Deloitte University Press essay and also concludes that audits are going to get “much more continuous, analytical, and at least semi-automated” but humans are still an irreplaceable part of the equation.

Here’s some areas where AI has the potential to blow up the status quo and force auditors out of their comfort zones:

Mine big data better

KPMG plans to utilize IBM Watson to:

analyze large volumes of structured and unstructured data related to a company’s financial information as auditors “teach” the technology to fine-tune those assessments, giving those teams faster access to precise measurements used to analyze anomalies.

This suggests that auditors will be able to increase sample sizes with less human labor. PwC's Assurance Transformation Leader Bill Brennan shared with Going Concern last fall that “in certain instances TODAY we are able to test 100% of a company's transactions.” Sampling may already be a thing of the past. And if it isn’t yet, it will be soon.

Boost audit quality

Isn’t it in the profession's best interest to boost audit quality? Deloitte’s Chief Innovation Officer Jon Raphael tells CFO:

Such technologies can enable auditors to automate tasks that have been conducted manually for decades, such as counting inventories or processing confirmation responses. And as a result, auditors can be liberated to focus on enhancing quality by evaluating advanced analytics, spending more time exercising their professional judgment, and providing greater insights.

With the added attention to detail it might even warrant a higher bill rate (cha-ching). In addition, most people would agree that providing more value to clients is better for the auditing profession too.

Force auditors to explore hyper specialization

Generic and relatively simple attestation and compliance tasks are turning into commodities. The cheapest one will win — AI-enabled or not. The trick for audit professionals (especially if you want a long and lucrative career) is to stay relevant by specializing in higher level advisory or consulting services — although we will need to grapple with that whole auditor independence thing again since it’s frowned upon to get too in the weeds with advisory while still doing audit. I’ll just say it… maybe it’s time to shift out of plain jane financial statement auditing?

Alternatively, auditors will need to ratchet down their audit services to a specialty. This involves creating niche service offerings that are so unique that programming a machine to handle them would be too expensive.

Deloitte’s cognitive technologies course instructor David Schatsky uses a good example:

A very wealthy man in Boston who is a broker of Dunkin’ Donuts franchises.  All he does it connect buyers and sellers of them.  Now we could probably automate that matching process but it’s so rare and so narrow that nobody would even bother.

Auditors who avoid commodity-like servcies will keep AI from gobbling up their audit jobs entirely, while still relieving themselves of some of the tasks auditors don’t want to do. Job security is one of the hallmarks of the profession, after all.

At the end of the day, I refuse to believe that cognitive technologies will come close to eliminating the auditing profession. Especially not during our lifetime. Maybe we just won’t need as many interns.

What do you think? Are we doomed or do we just need to act as pioneers in this new technological frontier?

Image: Someecards