online gambling singapore online gambling singapore online slot malaysia online slot malaysia mega888 malaysia slot gacor live casino malaysia online betting malaysia mega888 mega888 mega888 mega888 mega888 mega888 mega888 mega888 mega888 Artificial Intelligence Levels-Up In Accounts Payable

摘要: Artificial intelligence (AI) continues to make its way throughout the back office, exploring new areas to automate and optimize, freeing up human experts for more strategic initiatives.

 


images/process.jpg

▲圖片標題(來源: PYMNTS.com)

Today, the standard level of AI in a process like accounts payable (AP) most often appears as a technology that can automatically “read” key documents like invoices and identify and extract important information like supplier and pricing.

Alexander T. Hagerup, co-founder and CEO of Vic.ai, recently told PYMNTS that AI has a far greater opportunity in the enterprise, including in AP automation. In an interview, Hagerup discussed his vision for the technology’s presence in a notoriously complex and manual area of the back office and considered what the adoption curve may look like as more professionals feel comfortable with technology taking over important actions.

Beyond The Low-Hanging Fruit

As AI technology becomes more sophisticated, some of the most ingrained pain points of AP workflows no longer seem like difficult challenges to overcome. Hagerup noted that without intense configuration or ongoing maintenance, AI solutions can facilitate data extraction, for instance.

While described as the “low-hanging fruit,” his accomplishment is nothing to scoff at. Indeed, AI is now able to handle changes in workflow, like a new invoice template that a supplier uses, or identify deviations in typical patterns, like if per hour costs have risen.

“That’s massively powerful when you’re talking with large enterprises with millions of invoices and tens of thousands of vendors,” Hagerup said. “AI is just so superior.”

But Hagerup has grander visions for AI’s impact in accounts payable, and as the technology evolves, there are troves of other opportunities to capture and friction points to ease. They will, however, be more challenging to solve.

“What is really hard is accounting decisions,” he said. “That’s much more specific for each country, each region, each industry you’re in. You need, for the most part, educated accountants to make good accounting decisions.”

AI has even more ground to cover, though, with higher-level actions in its path.

Take, for instance, the actual act of paying a supplier. B2B payments aren’t as straightforward as hitting a “pay” button and moving on. Organizations need to have payments approved and ensure the accurate timing and method of payment while also prioritizing accuracy in case of any discrepancies or changes on an invoice. Then there are issues like duplicate payments and fraud to address too.

And, in order to truly automate and optimize B2B payments, those transactions must be made in relation to an organization’s cash flow. This, noted Hagerup, is like the “number one reason” why organizations keep that manual finger on the B2B payments trigger.

Embedding Intelligence Into AP

Vic.ai continues to make progress on these fronts and on other workflows within the back office. The firm recently secured a partnership with Botkeeper — Hagerup noted that such integrations and collaborations are win-win scenarios allowing Vic.ai to gain traction while elevating the functionality of partner platforms.

As AI continues to seep into more areas of AP workflows, the opportunity for firms to extract even greater value from the technology will only increase.

“The other piece that is often overlooked, but so important, is the actual intelligence from the data,” said Hagerup, noting that real-time cost optimization strategies, peer benchmarking and other analytical insights will be a top priority for Vic.ai moving forward.

Yet, even as AI explores the various challenging friction points of workflows, like AP, an even more existential hurdle exists that the tool must overcome: distrust.

It’s one thing for a technology to automatically extract data. But when an AI-powered solution begins making strategic decisions about when and how to pay a supplier, an organization might struggle to let go of the reins.

Hagerup likened this obstacle to the rise of autonomous vehicles. Today, it's relatively common to drive a car with some level of self-driving capabilities, though humans must still retain control of the wheel and be ready to take over at any moment. As the technology evolves, however, humans will begin to feel more comfortable with entrusting that vehicle with self-operation.

It’s a similar experience when businesses implement AI for the first time, Hagerup said.

“Maybe after three to six months of the car driving effortlessly and doing everything right, maybe you can kick back and trust the car more,” said Hagerup. “It goes hand in hand. [As] the AP team gets comfortable with the technology [and] they’re reviewing things ... then eventually, they don’t need to review [anymore]. They see everything is right ... There definitely is that trust element.”

轉貼自: PYMNTS.com

若喜歡本文,請關注我們的臉書 Please Like our Facebook Page:    Big Data In Finance

 


留下你的回應

以訪客張貼回應

0
  • 找不到回應