Reporter’s Notebook: Bill McDermott’s three big bets on the enterprise AI era

Hello and welcome to In The Channel from ChannelBuzz.ca, bringing news and information to the Canadian IT channel community for the last sixteen years. I’m Robert Dutt, editor of ChannelBuzz.ca, and your host for the show.

A rare double-up on In The Channel today. We’ve already dropped a great interview with Tim Coach from Cynomi on third party risk management. But this is something different altogether. 

This is a what I like to call a Reporter’s Notebook episode – no guest, just me and some thoughts I want to share from my time on the ground at ServiceNow’s Knowledge 2026 conference in Las Vegas. 

Specifically,  I spent about 40 minutes in a room with ServiceNow chairman and CEO Bill McDermott, along with my peers in the indsutry press, in a fireside chat format. You know, I’m always somehow just a tiny bit disappointed when there’s not an actual fireplace. But I digress.

McDermott is one of the most practiced and polished executives in enterprise technology – he knows exactly what he’s doing when he talks to press. But he’s also genuinely quotable in a way that a lot of enterprise CEOs aren’t, and I came out of that session with a few things I wanted to think out loud about.

Fair warning: he spent a meaningful amount of time on the ServiceNow valuation story and the disconnect between where Wall Street has the stock and where he thinks the business is going. I’m going to mostly skip that part. You’re solution providers, not analysts or institutional investors. What I want to talk about is what he said about where the industry is going, because I thought it was worth unpacking.

So. The line that’s stuck with me since I walked out of that room.

“Governance isn’t a feature. It’s the whole ball game.”

That’s the sentence I’d put on the poster if I were running ServiceNow’s marketing right now. And it’s not just a pithy line – it’s the entire strategic argument the company is making, distilled down to nine words.

Here’s the bet McDermott is making, and I think it’s worth understanding because it has real implications for how you think about the next few years of your business.

The first bet is that the AI race isn’t going to be won by the best model. It’s going to be won by whoever can make AI safe enough, governed enough, and deterministic enough to actually run an enterprise on.

He drew a distinction that I thought was clarifying. He said – and I’m quoting – “You can’t have a probabilistic solution for an enterprise. It has to be deterministic and it has to be right every time.”

That’s the core argument against the pure large language model play. An LLM gives you the best answer it can given the data it has. It’s probabilistic by nature. That’s fine for a lot of things. It is not fine when you’re running IT service management for a bank, or HR workflows for a public sector organization, or customer operations for a telco.

ServiceNow’s argument is that the governance layer – what they’re now calling the AI Control Tower – is the thing that makes AI enterprise-safe. Not just a nice add-on. The precondition.

The second bet is that AI isn’t optional. It’s arithmetic.

McDermott came back to this a few times. There’s a projected shortage of 50 million workers globally by 2030. The workforce isn’t growing fast enough to meet demand. Birth rates are declining. The enterprise can’t staff its way out of the problem it’s about to have.

And so the argument isn’t “AI will help you be more efficient.” The argument is “AI is the only answer to a math problem that is already in motion.”

He put it bluntly: AI isn’t coming for your job. AI is coming to do the jobs there won’t be enough people to fill. That’s a different pitch, and for a lot of your customers, it’s a more honest one. Is it a glass-half-full take? Sure. But considering the number of glass-completely-empty takes around AI and what it may do to the workforce of the future, I think it’s worth considering.

The third bet is the one that I find most interesting, and it was stated less explicitly, but I think it’s the most important one for your business.

The bet is that ServiceNow was built for this moment. That the platform that’s been processing workflow for twenty years – the one that already has a hundred billion flows running, most of them untagged and unidentified – was always the foundation that AI needed to land on.

He said it directly at one point: “This platform was always waiting for AI.”

And I think what he’s really saying is: the hard part isn’t the AI. The hard part is knowing what to do with it. Knowing which workflows to activate. Knowing how to govern what you activate. Knowing how to quantify the outcome. And ServiceNow’s argument is that twenty years of enterprise workflow data, and the relationships and trust that come with it, is a moat that a hyperscaler or a pure-play AI company cannot replicate.

There was a moment in the session – a little lighter – where McDermott talked about his relationship with Jensen Huang. He joked that every time he appears on stage with Jensen, NVIDIA’s market cap goes up by about a trillion dollars. He then pointed out that his own company’s multiple hasn’t quite kept pace with that. He was being self-deprecating in the way powerful people can afford to be. But the point underneath it was real: the NVIDIA partnership gives ServiceNow something the pure platform story couldn’t – a direct line into the AI infrastructure conversation, not just the AI governance conversation.

There was one stat he dropped that I’ve been chewing on since then, a statisitical representation of the challenge that he faces, and that you face. And also of the opportunity, for those who play the game wisely.

Only one in ten enterprises, by McDermott’s accounting, have actually moved from AI experimentation into AI that has genuinely impacted a core business process with real agentic workflows.

One in ten.

This is the CEO of the company that just staked its entire conference on the theme of “Welcome to Agentic Business” telling a room full of press that nine out of ten enterprises aren’t there yet. He wasn’t being pessimistic – he was making the case for the runway. But I thought it was an unusually honest thing to say out loud, and it’s worth noting.

Because if one in ten is the number, then for most of your customers, the “agentic business” isn’t something they’re navigating. But it’s probably something they’re aspiring to. And the opportunity for the channel isn’t to tell them about it. It’s to get them there.

And that’s exactly what the Go Live AI guarantee, and the AI Control Tower, and the whole machinery of what ServiceNow announced this week is designed to do. Give the channel a way to close the gap between the aspiration and the reality, at a predictable pace, with a quantifiable outcome. If you want a bit more on the Go Live AI guarantee and the AI Control Tower, we covered it in this morning’s episode of The Buzz, and tune in right here tomorrow, because ServiceNow’s channel leader, Michael Park, has a lot to say about the mechanices of the Go Live AI guarantee in particular.

One last quote I’ll leave you with. Someone in the room asked McDermott how he stays energized given the complexity of everything happening right now. He didn’t hesitate.

“This is the best time I’ve ever seen for innovation in the enterprise.”

He’s a CEO, so you take that with appropriate seasoning. But I was in the room, and I’ll tell you – to me it felt like he meant it.

More from Knowledge 2026 coming this week, so keep your favorite podcast app nearby.

If you’re finding In The Channel useful, please follow or subscribe wherever you’re listening – we’re on Apple Podcasts, Spotify, YouTube, and most major directories. Ratings and reviews are always welcome.

Until next time, I’m Robert Dutt for ChannelBuzz.ca, and I’ll see you in the channel.