Inside the Mind of an Intrapreneur: What It Really Takes to Drive AI Adoption

TL;DR: True AI adoption happens when people and technology meet halfway. Gil Neulander, the AI adoption lead at Activefence, shares his view on bridging the gap between people and technology, paving both roads at once while navigating the messy balance between innovation and security, enthusiasm and fear, vision and ROI.

What’s an “Intrapreneur” Anyway?

A few months ago, I came across a word that stuck with me: Intrapreneur.
It’s a twist on “entrepreneur,” but instead of starting something new from scratch, an intra-preneur builds innovation inside an existing organization.

That definition hit home. My official title is Director of Operations, but over the past six months, I’ve also taken on a new hat,  leading internal AI innovation and enablement at ActiveFence. My job sits at the intersection of people, process, and technology: helping teams experiment with AI responsibly, spotting where automation can remove friction, and making sure every initiative actually ties back to business value.

In practice, that means everything from defining our internal AI roadmap to hands-on experimentation; running workshops, enabling builders, measuring ROI, and building bridges between engineering, security, and operations. I act as an internal connector: translating between technical teams and business goals, helping each side understand how AI can make their work faster, smarter, and safer.

Since stepping into this role, we’ve built several internal AI tools across the company, from RAG-based knowledge assistants that make information accessible to everyone, to HR automations that improve how we source and screen candidates, to content generators that support our marketing and GTM teams.

Each of these projects reinforced the same truth: AI adoption isn’t about convincing people to use new tools; it’s about making those tools genuinely useful.

Paving Two Roads: The Real Work of AI Adoption

Driving AI adoption is a two-way street.
On one side, you’re helping people feel comfortable experimenting, overcoming fear, building trust, and showing them that AI can actually make their work easier.
On the other, you’re making the technology itself ready for them, secure, accessible, and integrated into real workflows.

You’re essentially paving two roads at once:

• From people to technology, by enabling experimentation and building confidence.
• From technology to people, by tailoring solutions that truly fit their needs.

The sweet spot is where those two roads meet. When AI stops feeling like “a new tool” and simply becomes part of how work gets done.

The Challenges Along the Way

1. Being Both Strategist and Executioner

One of the hardest parts of this role is that you’re both the visionary and the builder;  the one drawing the map and paving the road.

That means switching between high-level strategy and hands-on experimentation daily, and doing it while bringing others along with you. There’s no playbook for this kind of work. You’re paving an unpaved road, one small experiment at a time.

2. Balancing Security and Innovation

Yes, this might sound more like a CISO’s headache, but we feel it in Operations, too.

We work with some of the largest enterprises in the world, and we hold parts of their most sensitive data. It’s not even ours, which makes it feel twice as heavy.

The thought of that data falling into the wrong hands is terrifying. And the risk becomes even more real when you’re building “one source of truth” systems, those internal repositories meant to make company information easily accessible to employees. Suddenly, the same thing that empowers people can also expose us if we’re not careful.

Security is always at the back of my mind, even when we’re brainstorming something as simple as a chatbot. Every innovation decision has to happen alongside a policy conversation. It’s a tough balance, making things easier without making them riskier.

3. Helping People Cross the Fear Barrier

The fear barrier is real, but I’ve learned it’s not just fear. A lot of it comes down to habit. People have been doing things a certain way for years, and it’s hard to convince them to change what already “works.”

AI can feel intimidating. Some worry about getting things wrong; others worry about what it means for their role. But most often, people just don’t know where to start. They’ve built workflows and shortcuts over time, and asking them to rewire that overnight feels like asking them to learn a new language.

Part of my job is helping people cross that psychological gap, showing that AI isn’t here to replace what they do, but to make the boring parts disappear so they can focus on the work that actually matters.

3. Staying Afloat in the Constant Flow of Innovation

Every week, someone sends me a new AI tool they’ve just discovered. “You have to see this, it’s incredible.” And they’re usually right.  The pace of innovation is relentless, and the hype cycle never sleeps.

The challenge isn’t curiosity, it’s prioritization. You can’t test everything, and not every shiny tool meets enterprise standards for data handling, compliance, or reliability. But when everyone wants in, you need a clear way to evaluate what’s worth exploring and what’s just noise.

The trick is to keep the excitement alive while steering the energy toward tools that are actually enterprise-grade: secure, scalable, and relevant to our needs.

5. Measuring ROI in a World of Unknowns

Innovation always sounds exciting until someone asks, “So, what’s the ROI?”

When you replace a manual process with AI, that’s easy to calculate. But when you’re inventing something completely new, like automating a process that never existed before, it’s harder to put a number on its value.

A lot of this work involves making educated assumptions and asking for budgets before the proof exists. It’s uncomfortable but necessary. Over time, the value becomes clearer: time saved, fewer bottlenecks, and smoother handoffs.

Still, you need to have the confidence to bet on ideas that don’t yet have a metric attached to them. That’s what separates an experiment from a true innovation effort.

Building the Bridges: What’s Worked So Far

If there’s one thing I’ve learned, it’s that adoption doesn’t happen because you announce a new strategy; it happens because people experience small wins that feel real.

One thing that really helped us kick-start momentum was the AI hackathon. It’s not a groundbreaking idea; plenty of companies do them, but when you have leadership that backs it and treats it as a culture-setting event rather than just a few smiling photos for the company’s social media, it actually works.

It wasn’t just about the prototypes we built; it was about tone-setting. That day showed people that AI isn’t just trendy, it’s something they can play with, shape, and use. It also sparked a wave of follow-up initiatives, like the internal learning spaces we’ve since built to help employees keep exploring on their own.

Another big enabler has been cross-department collaboration. Every time we run a learning session, we bring together builders, designers, and the people who actually feel the pain points, those who live the problem. That mix is where we create tools that truly move the needle.

And on a personal level, this collaboration is what keeps things real. I work closely with our CISO’s office to assess whether tools are safe, with Finance to prove value and evaluate budgets, and with Ops and Product teams to make sure our efforts stay connected to real workflows.

For me, this kind of collaboration is where the real culture shift happens. It turns AI from a side project into something everyone has a stake in improving.

What I’ve Learned (So Far): Takeaways for Other Intrapreneurs

Six months isn’t a long time, but in AI time, it feels like a lifetime. Things change fast, the tech, the tools, even the expectations. What’s consistent, though, are the lessons that come up again and again.

Here’s what I’ve learned so far:

1. Start Small, Celebrate the Wins.
   Don’t wait for the big success story. A single small project that saves a team time or cuts a repetitive task can build more trust than any roadmap presentation. Small wins compound fast.
2. Build Together.
   The best ideas come from collaboration, not just between technical and non-technical people, but between those who build and those who feel the pain. When both sit at the same table, solutions tend to be simpler, sharper, and easier to adopt.
3. Use What Exists.
   There’s no glory in reinventing the wheel. Open-source tools, templates, and shared prompts are incredible accelerators. The real value comes from adapting, not building everything from scratch.
4. Stay Curious and Connected.
   The AI landscape evolves daily. I make it a point to stay close to online communities and peer networks,  to see what others are trying, what’s working, and what’s not. You can save weeks of trial and error just by learning from someone else’s story.
5. Measure What You Can, Learn from What You Can’t.
   ROI isn’t always obvious. Sometimes the success metric isn’t money saved, but time, creativity, or culture gained. The more we track those intangible wins, the easier it becomes to justify the next experiment.

The Road Ahead

The longer I do this, the more I realize that driving AI adoption isn’t a one-time rollout; it’s a living process. You don’t “finish” building with AI; you learn, adapt, and evolve alongside it.

Every experiment (the ones that succeed and the ones that don’t) teaches something about how people and technology can work better together. What starts as a small win gradually becomes part of the company’s DNA: the instinct to test, to learn, to improve.

In the end, the real measure of success isn’t how many tools we build, but how naturally and responsibly AI becomes woven into everyday work. Safe, secure, and genuinely useful AI makes innovation last.

—

If you’re thinking about how to bring that mindset into your own organization, talk to our team about building AI programs that empower teams while keeping security and responsibility at the core.