Four types of debt are holding companies back from unlocking AI value

Like the lucky mystery briefcase on Deal or No Deal, enterprise debt is standing between global companies and a large sum of money.

According to a new report from Genpact in partnership with HFS Research, 85% of companies admit that enterprise debt is “actively limiting” the value they can extract from AI by creating an unstable foundation for such projects to succeed.

The report, based on a study of 2,002 global enterprise executives from 16 industries, concluded that a cumulative total of about $18 trillion in “potential value” is left on the table for Global 2000 companies as a result of unaddressed debt. (Genpact calculated this value by using estimates from respondents of how much money they believed their organization would save by addressing each enterprise debt to estimate impact across all Global 2000 companies.)

What exactly is enterprise debt? Genpact Global Agentic AI Officer Vijay Vijayasankar told IT Brew that four kinds of debt are currently throwing a wrench in companies’ AI endeavors:

• Technology debt: A well-known and unwanted friend for a number of enterprises, Genpact said development teams spend more than 40% of their time tending to existing technology debt, taking away time from building out new capabilities and catering to transformation initiatives. Technical debt remains a “crushing problem” for organizations because companies won’t shell out funds to upkeep technology until it becomes a business risk, according to Vijayasankar.
• Process debt: Costs derived from “inefficient or manual processes”. According to the report, employees spend around 40% of an average week addressing outdated processes.
• Data debt: This is the misalignment between the quality of organizations’ data and what they need for AI. IT Brew has previously reported how fragmented and poor quality data can become a bottleneck for AI projects.
• Talent debt: This includes the skills gap between the current workforce and what is required by “human-agent operating model AI [tools].” Genpact estimates that only around one-third of the workforce is “AI-ready.”

Vijayasankar described the cumulative debt total as a result of “things that could be done better, faster, [and] cheaper.” In order for organizations to reap the full benefits of AI, he added, they must work to dismantle their enterprise debt.

How to tackle the debt crisis. While debt remains a large problem for organizations trying to enact AI transformation, Genpact said many have no funded plans to address it.

Vijayasankar suggests companies conquer enterprise debt by figuring out which areas need to be prioritized, adding he doesn’t recommend putting an “equal investment” into fixing all four types of debt.

“Yes, all four of them need to be addressed, but not all four of them are equally problematic in every company,” Vijayasankar said. “It’s a business case-based decision.”

Companies should also reject the misconception that enterprise debt is a one-and-done deal, he added.

“My closet is a great example of why that theory doesn’t hold true,” Vijayasankar said. “The next time I have laundry, and I don’t have time, I just toss the whole thing into that room, and then seven days later I can’t walk into the closet, and that’s when I start folding. This kind of an approach is very common in enterprises today.”

Finally, Vijayasankar said tackling enterprise debt requires a lot of “management guts,” and that it is up to leaders to frame eliminating debt as a worthwhile priority for their business: “It’s a leadership crisis, more than anything else.”