From ChatGPT has exploded into the zeitgeist In late 2022, the business buzz around this topic has been filled with sometimes exaggerated claims about the extent of the role that generative AI will play, including dire warnings to financial executives about the AI skills they need. need to keep their jobs.
But what AI skills exactly do finance leaders need, and how can they develop them? CFOs and other finance leaders may be happy to know that these skills don’t mean learning to code your own AI models. However, this involves understanding how AI works and what it can do, said the KPMG managing director. Joseph Parente, American leader in AI consulting.
“You don’t need to be a software engineer or even know how to write Python code. But you need to understand the concepts of machine learning,” Parente said in an interview. “How do machine learning and natural language processing actually work? What is possible, what is not?
Octavio “OJ” Laos, director of AI at California-based accounting firm Armanino, has sometimes been surprised by financial executives’ misconceptions, so he recommended trying some of the generative AI models available first. to the public and often free. so they can understand how it works in real life.
“What they need more is just real interaction with him, maybe not so much knowledge as experience. It’s hard to conceptualize what’s at stake until you use some of these tools,” he said in an interview.
CFOs must link AI to goals
For the CFO in particular, they need to learn how to connect AI to their organization’s overall strategic goals, and a lot of that comes down to finding good use cases that fit their individual situation, according to W. Michael Hsu, CEO and founder of DeepSky of California, an outsourced CFO consulting and accounting firm.
“For CFOs, AI becomes a sandbox for strategic thinking. The role of the CFO is no longer just about managing financial data; it’s about collaborating with AI to explore new models, quickly test ideas and make better decisions faster,” he said.
To that end, a CFO may need to learn how to leverage AI to interpret data and detect anomalies, or model scenarios to test strategies and ideas, Mr. Hsu said.
KPMG’s Parente also emphasized that CFOs should also focus on finding compelling use cases before deploying AI, noting that one CFO he spoke with wanted to implement a solution in the entire company, but had not proposed specific applications. He likened it to a CFO of an agricultural company who had a more specific plan to use technology to determine whether a cow was being milked correctly.
CFOs must also put AI products through rigorous budget testing to determine if they are worth their cost and fit into their budget.
“What is the true cost of this? Everyone thinks about the benefits: can I reduce my work, improve my quality. But… what is the cost to acquire it, build it, maintain it? That’s the responsibility of a financial executive,” Parente said.
Controllers to examine specific applications
In contrast, someone at the controller, treasurer or CAO level is more grassroots and so their AI skills should revolve around the specific needs of their organization and how they could be improved, a said Armanino’s Laos.
“It’s not so much about planning as it is about looking at practical applications,” he said to executives in these positions.
At this level, AI skills focus more on specific applications, such as using AI to automate compliance and reporting, error detection and cash flow forecasting, DeepSky’s Hsu said. Examining AI use cases in financial services could force controllers and treasurers to think about how AI could disrupt segregation of duties, Laos said.
A fundamental principle of a sound system of financial controls, separation of duties includes systems set up so that no single person is responsible for all stages of the financial process, which can range from payment authorization and approvals to reconciliation, as defined by the University of Oxford .
“Traditionally, roles such as controller and treasurer have been separated to avoid fraud,” Laos said. “A question worth exploring is whether AI can (serve) dual purposes without human bias or risk of fraud.”
Become competent
Developing AI skills is a combination of learning by doing, consuming content about AI, and finding learning communities or events around AI, experts said.
Both Armanino’s Laos and KPMG’s Parente were skeptical of generic AI classes, which tend to only scratch the surface. Instead, Parente recommends role-specific training that focuses on particular positions like CFO, CEO or treasurer.
Large accounting firms such as KPMG partner with various universities to provide such targeted training. But no matter what one decides to do, Parente said the important thing is to go for it and not be intimidated, noting that, anecdotally, some leaders’ understanding of technology remains relatively “light » because they fear its complexity.
“There’s this stigma…but at some point it becomes business as usual,” Parente said. “We all learned spreadsheets, we all learned financial systems, all of those things came over time. It will happen again. It’s the start of a journey, but it’s not the first time you’ve done it.