Artificial intelligence is bringing plenty of benefits to finance, from accuracy on the balance sheet to stronger predictions in the supply chain. And for chief financial officers, AI offers them something invaluable: more time in their day.
When used to automate repetitive and manual tasks, such as data entry and invoice processing, AI can open up opportunities for finance leaders to focus on financial planning, analysis, and more strategic work, said Joseph Wolk, executive vice president and CFO at Johnson & Johnson.
In a conversation with MIT Sloan economist at the 21st annual MIT Sloan CFO Summit, Wolk said that AI is poised to “free up time for people in the finance function to be better influencers of decisions.”
“At Johnson & Johnson, and in the finance function specifically, we have a little mantra: We’re business leaders with a finance expertise,” Wolk said. “The finance expertise is secondary to helping the business solve problems.”
Regarding the ways in which AI has swiftly gained worldwide acceptance, Johnson attributed AI’s adoption to its “fun element” and existing infrastructure.
“We’ve already built digital rails that can distribute this product instantly everywhere,” Johnson said. “You might need to scale up a few things in terms of bandwidth, but mostly not. Mostly it’s available right now.”
Johnson is the co-author of “Power and Progress: Our 1,000-Year Struggle Over Technology and Prosperity,” which explores the history of AI as well as other historical technological transformations.
“I think AI presents a real opportunity to do something different — to run your finance function better, to help your companies invest and make more money [and] make the shareholders happy, and to generate more good jobs,” Johnson said.
In their talk, Johnson and Wolk identified four use cases for AI in finance:
Predict supply chain disruptions in real time
Artificial intelligence can already be used to streamline kinks in the supply chain across manufacturing, distribution, and delivery. Wolk believes that in the future, AI will help more with prediction. This will be invaluable in health care, he said, where having a responsive supply chain can be “lifesaving.”
“Where I’d like to see AI evolve for us is in getting much more into the predictive side of things,” Wolk said. One example: using predictive analytics to identify and anticipate potential disruptions in the supply chain in real time.
“We should know when a medical device leaves the supply room from a hospital, or a drug or a biologic leaves the pharmacy [and] needs to be replaced,” he said. “We should know that [in] real time; those are some of the great, great benefits.”
Lower the cost of drug development
Wolk said it’s possible that AI could lower the cost of drug development by improving the chances of matching the right patient with the right clinical trial. In health care, researchers are already using machine learning algorithms to identify existing drugs that might have the potential to treat different diseases.
“If you think about what these algorithms can do in a very short period of time relative to where we were just four or five years ago — you’re going to match up and recruit for clinical trials with patients who are likely to respond to that type of treatment,” Wolk said. Although the practice is not yet widespread, “I think that’s the potential,” he said.
Ease the burden of compliance
Across different locations, AI is helping J&J stay compliant and better manage its legacy systems by providing better integration and “greater surety on the financial reports that you’re signing off on,” Wolk said.
J&J does business in 170 countries worldwide, which can present some compliance challenges, he said, adding, “You can imagine the different type of accounting that goes on.” AI is helping the company clean up its data and make “great progress around data integrity.”
Augment human capabilities
MIT Sloan’s Johnson said that there’s a lot of “really interesting potential” to use technology to augment human capabilities, particularly for workers who don’t have a lot of formal education.
He cited joint work at Stanford and MIT that showed that contact center agents with access to a conversational assistant experienced a 14% boost in productivity on average, with the largest gains impacting new or low-skilled workers. “Your customer service representatives who are less experienced become better,” Johnson said. “They like their job better. The people that they’re helping are more satisfied, and you get less turnover in those positions.”
While hailing the potential benefits of AI, Wolk reiterated his belief that humans will always be crucial to the health care industry. “You can diagnose things [with AI], but unless you have somebody looking over that medical chart, something could get missed.” In other words, there is still a lot of progress to be made, Wolk said.