By Stephanie Walden, Contributor
To business leaders, emerging technologies are often beacons that represent coveted first-mover advantage. Artificial intelligence (AI) is proving to be no exception. According to a recent Forbes Insights survey of 700 C-suite executives, a whopping 77 percent of business leaders say their organizations are either planning or piloting AI within their organizations. Nearly three-quarters (74 percent) have 10 or more AI initiatives currently underway.
While these numbers show the enthusiasm around AI’s potential, they also indicate potential fragmentation in its adoption. This lack of a clear, enterprise-wide strategy comes as no surprise to Dell EMC President and Chief Technology Officer John Roese. “When we entered the digital era, we didn’t enter it—in most enterprises—with a holistic strategy,” he said.
“…we all know that data is the fuel of the enterprise.”
—John Roese, Dell EMC President and Chief Technology Officer
While going after the “bright, shiny object” was the right move at the time for many organizations to claim first-mover advantage, he said, now is the time to reverse some of the fragmentation and create a more mature, comprehensive approach to how we deal with our data in the long-term “because we all know that data is the fuel of the enterprise.”
So, what is standing in the way of a more comprehensive strategy to AI adoption? Here, several experts weigh in.
The State of Data Readiness
When queried about barriers to AI adoption, executives were evenly split in their responses, citing themes like a lack of data literacy and AI expertise, challenges with data management and security, difficulty balancing budgets and measuring ROI, and antiquated IT infrastructure and tools.
Upon further investigation, however, another factor appears to be responsible for the sluggish pace of AI adoption. The study found that business leaders admit a lack of system centralization and accessibility. In fact, only 12 percent of respondents reported having an enterprise-wide data strategy. What’s more, a full 80 percent noted that 40 percent or less of their company’s data is readily available across all teams. In short, data readiness is lacking.
It’s understandable how this predicament came to pass. When the term “data science” first became a hot boardroom topic, many executives challenged their best and brightest teams to dream up innovative programs using the new, promising tool—and it worked. Customer service teams fine-tuned loyalty programs. Employees working on supply chain optimization developed advanced strategies to boost efficiencies. Engineering teams began experimenting with data-driven products.
But as a result, these segmented processes of innovation—with little or no overarching strategy in sight—led to data silos. “We created a data fragmentation issue,” explained Roese. “But the real value of [the data era] is the ability to correlate data across domains—to be able to understand the relationship between your supply chain and your customer relationship and your engineering environment. ”
Now that companies are aware of this problem, they’re making moves to tweak data systems so that they’re more cohesive. Tamara McCleary, CEO of Thulium, a social media analytics and consulting agency, is optimistic that silos are starting to dissolve and merge. “Even though most organizations have siloed platforms today—cyber security over here, customer experience over there—I believe we’re moving very rapidly toward a time when it’s all going to be integrated,” she said. “Customer experience is going to be intertwined with the security aspect of AI running in the background. It will all be seamlessly integrated, so it isn’t separate as it is today.”
“Customer experience is going to be intertwined with the security aspect of AI running in the background. It will all be seamlessly integrated, so it isn’t separate as it is today.” —Tamara McCleary, CEO of Thulium
Other Barriers: Talent and Culture
In addition to the lack of an enterprise-wide data strategy, experts call out gaps in a comprehensive people strategy as another hurdle to enterprise-level AI. For starters, Mike Crones, CIO of Draper, a not-for-profit research and development organization, points to the oft-lamented talent shortage of skilled, business-minded data professionals. “Skill sets within the workforce are going to be as great of an impact to the success of a data strategy as access to and presentation of the data itself,” he said. “Without insightful and skilled professionals, business leaders will become frustrated with the inability to leverage meaningful data.”
Along this same vein, company culture is another crucial factor when it comes to AI readiness. Daniel Newman, principal analyst of Futurum, a strategic research and analytics firm, points out that given the speed of digital proliferation, many companies’ cultures aren’t ready to adopt new technologies—particularly if executives are vague in their explanations of the technologies’ long-term functions. “Areas like automation create a lot of fear, and that fear tends to slow down the rate in which people buy into change,” he said. “If the companies and the people in the companies don’t buy into change, new technologies are never adopted at the rate they should be.”
In order to begin breaking down these barriers, Roese suggests a strategy overhaul that prioritizes interconnectedness and data transparency across all tiers of the organization. He notes that business leaders should strive to set up their teams and communication processes to facilitate “big breakthroughs,” which typically happen when data is accessible from multiple dimensions and across different domains.
“We have to have balance. Let’s not slow down the innovators, but now we have to put a comprehensive data strategy behind it,” he said. “We have to create a chief data officer, we have to understand where our data is, we have to curate that data, and then we have to look for opportunities to actually cross domains, which is where I think we’re going to unlock the significant value.”