Neither Panic, Nor Hype: Here’s A Better Way to Think About AI

I have made a conscious effort to avoid writing about artificial intelligence (AI).

Not because I don’t use it. I do.

Not because I think it’s overhyped. I don’t.

And certainly not because I think it’s going away.

I’ve avoided writing about AI, because I’m not an expert.

The statement to follow will probably annoy some people, and that’s okay. If we’re being honest, I suspect most of the people posting about AI every day aren’t experts either.

My social feeds remind me of the stretch between 2010 and 2014, when everyone suddenly became a social media expert. Back then, it felt like there were more social media gurus than there were businesses willing to advertise on platforms like Facebook. The irony was that the only way to become an expert was to spend real money, run real campaigns, make mistakes, and learn from them. Most people hadn’t done any of that.

Today the same thing is happening with AI. Every day I see posts declaring that ChatGPT is the only marketer you’ll ever need, that prompt engineering is dead, that AI agents are replacing entire departments. Someone somewhere always seems to be doing six years of work while sitting on the toilet.

The content reminds me of the early days of the app formerly known as Twitter. Remember those updates? Just woke up. Walking to work. Lunch goals. A lot of AI content feels remarkably similar: endless updates about the tool itself, rather than what anyone is actually accomplishing with it.

And yet, despite avoiding the conversation publicly, I can’t escape it professionally.

As a growth consultant, advisor, board member, and occasional sounding board for CEOs, I spend my days helping organizations figure out how to grow. I work with small businesses, nonprofits, and established companies across Canada, the United States, the United Kingdom, and the Middle East. But no matter the industry, geography, or size of the organization, the same topic keeps coming up: AI.

But it doesn’t come with excitement. Nor with confidence. Instead, it’s confusion.

For instance, the CEO wants to know where AI fits into the business strategy. The board wants to know whether the organization is falling behind. The leadership team wants to know which tools are worth investing in. Employees want to know if they’re training their replacement.

Everyone is looking for answers. But very few people seem confident in the questions they’re asking.

Here is the part that keeps surprising me. The two opposite reactions lead to exactly the same place.

One board I worked with became so concerned about the risks of AI that it effectively prohibited the technology across the organization. The intention was understandable. The outcome was paralysis.

I’ve seen the reverse just as often. Organizations become captivated by a single tool and decide it will solve everything. Teams spend months talking about AI, instead of figuring out how to improve customer acquisition, streamline operations, or generate revenue. The technology becomes the strategy. That’s usually when things start to go sideways.

Ban it and you freeze. Worship it and you drift. Both roads end in the same spot, which is a long way from the actual work.

One of the most interesting conversations I keep hearing revolves around critical thinking. It’s also one of the few concerns I genuinely share.

A leader recently told me, “I’m excited about what these tools can do, but I’m worried people will stop thinking for themselves.” It’s a fair concern. Like any muscle, thinking weakens when you stop using it.

I’ve watched junior employees submit work that is technically correct but strangely hollow. The words are there. The thinking isn’t. I’ve sat in interviews where candidates gave suspiciously polished answers after long pauses and quick glances off camera. Maybe they were consulting notes. Maybe they weren’t. But it left me wondering where the person’s thinking ended and the machine’s began.

The leaders who seem least worried about this are the ones who treat AI as a first draft rather than a final answer. They let it produce; then, they argue with it. They make their teams defend a recommendation the tool generated, not just forward it. The muscle stays strong when the machine has to earn agreement instead of being granted it.

At the same time, I’ve seen AI help people accomplish things they never could have done on their own. A small business owner with no design team can create marketing materials. A nonprofit can summarize hours of meeting notes in minutes. A founder can build a prototype without hiring a developer.

Those aren’t productivity gains. They’re capability gains. And that’s where things get interesting.

I recently came across a post suggesting we’ve moved from a creator economy to a builder economy, and it resonated with me. For years, technology helped us create content faster. Now, it helps us build faster—websites, applications, processes, workflows, ideas.

The question isn’t whether AI can help you build. (It can.) The question is what you’re trying to build in the first place. What problem are you solving for?

That’s where many organizations get stuck. They’re searching for best practices, and the problem is that there aren’t any. At least not yet. The technology is evolving too quickly. Every company is experimenting. Every industry is figuring it out in real time. There is no universal playbook. No one has all the answers. And anyone claiming they do should probably be treated with healthy skepticism.

What I do know is that every major technological shift creates two groups of people. Those who dismiss it, and those who believe it will solve everything. Both are usually wrong. Though it’s worth saying that the cost of being wrong isn’t evenly shared. A startup that ignores AI and a hospital that over-automates a clinical decision are not making the same-sized mistake. The right amount of caution depends on what happens when you’re wrong.

Take Klarna, the buy-now-pay-later company that made headlines when it leaned hard into AI and credited the technology with doing the work of around 700 customer service agents. The move was celebrated by some and feared by others as a glimpse into the future. But then, the company changed course. By 2025, its CEO publicly admitted the AI-first approach had produced lower-quality service, that cost had become too dominant a factor in the decision, and that Klarna was bringing human agents back into the mix. Complaints had climbed, and the hardest conversations, the disputes and the fraud claims, were the ones the bot handled the worst.

The lesson wasn’t that AI failed. The lesson was that people matter. Technology works best when it enhances human capability, not when it assumes it can replace it.

So, where does that leave the rest of us? Honestly, probably in the same place we’ve always been. Trying things. Making mistakes. Learning. Adjusting. Repeating.

I don’t think most organizations need a comprehensive AI strategy right now. I think they need curiosity. I think they need permission to experiment, which in practice means a leader saying out loud that a failed experiment is a cost of learning, not a mark against you. And I think they need to focus less on the tools, and more on the problems they’re trying to solve.

Most importantly, I think they need to remember that technology has always been the easy part.

The hard part is deciding what matters. The hard part is thinking. The hard part is building something worth improving.

And despite all the incredible advances we’ve seen, that still feels like a stubbornly human problem.