As AI Ambitions Stall, Enterprises Find the Fix in Data Governance Automation

For decades, companies have been accumulating vast stores of unstructured data. Today, that poorly managed foundation is threatening to derail their AI ambitions. Most leaders assume the solution is months or years away. But the tools to automate and unify data governance are already here.

Some believe that already, the race to enterprise AI is starting to slow down. After just a few years, a complex web of state regulations, industry standards, and outdated internal processes is limiting innovation. Now, some experts say the diminishing pace could be a data governance issue.

The industry is aware of the problem, according to Stephen Gatchell, Vice President of Data and AI Strategy at BigID. Having spent his career shaping data strategy at enterprise giants like Bose Corporation and Dell, Gatchell is a seasoned expert on the subject of governance architecture. Today, he also helps define industry-wide best practices as a special interest group lead for the EDM Council.

"Automated governance is already here," Gatchell says. "Not in the next six months, not in the next year. Right now." While leaders wait for a silver bullet, he continues, the tools to automate and unify governance are readily available today. In his experience, the bottleneck is more about mindset than anything else.

• Framework fundamentals: For Gatchell, the path out of the compliance maze begins with a radical reframing of the goal. Instead of trying to boil the ocean by managing dozens of frameworks like NIST and ISO in isolation, he recommends an automated, unified approach that focuses on the fundamentals. "If I do these 10 things, do I comply with 80% across all these different regulatory frameworks? Spoiler alert: the answer is yes. It's simple stuff, really. Inventory and classification, tagging and labeling your data, data ownership, and identifying business use cases to measure what good looks like."

• Data debates: The alternative is a state of perpetual "compliance chaos," Gatchell says. Here, the effort is decentralized and unsustainable, draining budgets and human capital. But even worse, it creates internal friction. "It's going to be the traditional data analytics conversation where you go into a room and somebody says, 'Here's our revenue and here's our bookings.' And then the chief financial officer says, 'No, that's not it.' And you start arguing about the data versus actually making business decisions."

The most mature organizations are already moving past this reactive posture, according to Gatchell. And their leaders are already engaging in more sophisticated conversations at the highest levels.

• Risk vs. reward: In a world of varying regional regulations and use-case sensitivities, these are the companies that see a one-size-fits-all approach to risk as naive, Gatchell says. "The companies that are smart about this already have a risk tolerance discussion, and they talk about how to balance the governance versus the speed of innovation. For instance, if you're in the medical field using AI to diagnose patients, there's a huge risk involved, and your risk tolerance is very minimal. That is much different than a retailer recommending a different pair of pants or shoes to an individual."

Ultimately, no tool or vendor can replace the need for strategic alignment and a fundamental shift in imagination, Gatchell concludes. Instead of running AI programs as disconnected pilots chasing efficiency gains, he suggests solving core business problems first. Pointing to the example of a salesperson, he explains how using AI to transcribe meeting notes is an "efficiency mindset." A "transformational mindset," meanwhile, asks AI to solve the entire problem.

For Gatchell, that shift from process to outcome is key. In closing, he offers a final takeaway for leaders looking to break the governance gridlock: "It's all about data. Understand the data management frameworks, pick one, and execute it. And use technology to do it. Absolutely, use technology to rethink the current process and improve it. Just rethink the whole outcome."