The race to adopt AI is well underway. Boardrooms are deep in conversations about models, tools and rollout timelines. But in the rush to move forward, something more fundamental is being overlooked.

Marc Potter sees it clearly. And in his view, it’s the difference between AI that looks impressive and AI that actually delivers.

“AI is prevalent as a topic, but so few people are talking about the things that really matter,” explains the CEO of Actian, the data intelligence company helping customers build, enrich and activate their metadata foundation for AI.

What really matters, in his view, is whether the data feeding the AI model can actually be trusted and whether the people using it feel equipped to act on what it tells them.

The ‘make or break’

The strategic thread running through Actian’s recent years – the acquisitions, platform advancements and investment in data observability – begins with a single conviction.

“Data has now become the ‘make or break’ on whether or not you’re successful with AI,” Potter says.

“Your data strategy is your AI strategy. Most organizations haven’t made that connection yet.”

Every move Actian has made is designed to let organizations drive their AI journey on their own terms, with their own data, governed the way they actually run their business. The key, he argues, is context – and it’s where most organizations are still falling short.

“Our message really is the importance of context,” he notes. “That’s what data is missing.”

Context, in Potter’s view, means giving every data asset a semantic identity: what it means in business terms, where it came from, who owns it and what policies govern it.

“Without context, AI agents aren’t reasoning,” he says. “They’re guessing.”

From data to decisions

The answer Actian is building toward is an intelligence layer specific to each organization, which reflects how they actually define their business rather than relying on generic assumptions.

“If you give your data context, you create an intelligence layer that is more relevant for your business and your decision-making than generic answers,” Potter points out.

Most organizations have data, but they are missing the ability to find it, trust it and put it to work. The Actian Data Intelligence Platform is built to give both people and AI agents a clear answer to three questions: How do we access the data? What does the data mean? And can it be trusted?

Answering the access question is where Actian AI Analyst, built on the acquisition of Wobby.ai, comes in. This capability built by a journalist, they discovered, translates directly into the trust problem at the heart of enterprise AI. That need for trust becomes even more apparent in how customers approach AI today.

Many arrive somewhere in the middle of their AI journey, Potter notes.

“They’re learning through trial and error,” he says. “But the goal is always the same.”

With Actian AI Analyst, a business user can ask a complex question of their data in plain language and get a trusted, accurate answer back in seconds.

“That’s how you move from data to decisions,” Potter reveals. “Instead of waiting for days for a new dashboard, anyone can get verified answers in real time with a clear audit trail.

“Ultimately, we’re trying to get customers to a point where more people across the business can make faster, more accurate decisions and act with confidence.”

Trust is the multiplier

Potter is quick to acknowledge that trusted data is only part of the equation. Another part is people who feel safe enough, equipped enough and genuinely trusted to act on what the data shows them.

“There is no wiggle room on this one,” he insists. “A happy employee will create a happy customer engagement.

“The best leaders will lead by example, learning and using the technology alongside their employees. I’m always looking for how technology can make others’ lives easier. I don’t want people to work more hours. I want them to work smarter. AI can be a way to do that.”

The principle is the same whether he’s talking about employees or customers: meet people where they are, build trust before expecting results and understand that change at this pace is genuinely challenging.

“We have to be empathetic that change is hard and new technology can be scary,” he acknowledges. “They’re watching the news, seeing people losing their jobs because of AI.”

Fix not fault

At the heart of both Potter’s leadership approach and Actian’s Data Intelligence Platform itself is a shared conviction.

“Building trust is the foundation of everything,” he insists. “When you can trust what you’re working with, you make better decisions.”

Potter has long believed that governing data and empowering people are expressions of the same belief. And that belief has a practical expression inside the company’s four walls – fix, not fault.

“If people realize that I care, I want them to collaborate and it’s OK to share the bad news – because it’s about the fix, not the fault – then we can take care of a problem before it is too detrimental,” he explains. “We can pivot and adjust.”

The parallel to data governance is not coincidental. Organizations that catch data quality issues at the source face a fraction of the cost and disruption of those that discover problems downstream.

“Actian’s governance by design approach is built on exactly that principle,” Potter says. “Don’t wait to clean up data after it has already caused damage. Set the governance rules early and enforce them in real time.”

The real competitive advantage

As AI accelerates, Potter believes many organizations are looking in the wrong place for what will set them apart.

“AI has shifted where competitive advantage actually lives,” he says.

“It’s no longer the code you write because AI can generate that in days. It’s your people’s ideas that set you apart. Ideas become products, and that’s still a distinctly human contribution.”

Harnessing those ideas at scale requires a data foundation that reflects the organization’s own business logic, definitions and context. Without the right governance infrastructure, sensitive data spreads without controls, ownership becomes unclear, compliance exposure grows and leaders end up making decisions on data they can’t fully trust.

For Potter, the answer isn’t to restrict AI but to build the data foundation that gives organizations control over how AI is deployed and what it can access.

“The first phase was just getting everyone using AI,” he explains. “To really scale it, you need clear policies on what data AI can access, what it can’t and how your people and AI can safely interact with it.”

Getting that foundation right, he believes, is what moves organizations from experimenting with AI to realizing real business value from it.

A legacy written in people

For a CEO who has spent his career at the intersection of technology and the humans who use it, legacy lands somewhere unexpected.

“I’ve always wanted to make a real difference in people’s lives. That’s what matters,” Potter says.

“If that’s what my legacy becomes, that I’ve made an impact on the people around me, then I’ll leave a happy man.”