In case you missed it, yesterday I wrote about an odd bit of timing in the news. 

On the same day U.S. diplomats were reportedly told to push back against foreign “data sovereignty” measures, U.S.-based OpenAI detailed how its platform had been used by a foreign law enforcement agency. 

Trade pressure on one side. Control over information on the other. This comes down to who keeps leverage when data moves across jurisdictions.

Turns out, isolationists can be surprisingly selective.

Looking back at some of our covers in the last few months, the intersection of AI, government, and very large sums of money is a recurring theme.

Microsoft’s $19-billion investment in Canadian AI infrastructure is an economic development story. Data centres, jobs, capacity. It also raised questions about digital sovereignty and who should shape the backbone of the systems the government increasingly relies on. 

At the same time, Ottawa recently completed consultations as part of developing a national AI strategy. The end of that phase came with some fanfare, and more than 11,000 submissions is nothing to turn your nose up at, but we’re still waiting on that strategy. 

Meanwhile, adoption isn’t waiting. 

In Edmonton, firms are working on AI tools designed to help speed up Canadian policy work, including drafting legislation. Faster drafting sounds like administrative modernization (Yay), as long as we don’t forget the value of debating why the legislation is being written. Maybe we’ll see parliamentary prompt debates in the future. 

The good news is that there are leaders focused on implementing AI in ways that help those who will use it adapt as it comes online.  

There are also reminders, like Nova Scotia’s digital health overhaul, that large institutions can use new technology to make services faster and easier to access. 

AI isn’t hovering outside the state waiting to be regulated. It is being financed, debated, integrated, and designed inside it. The money is real. The policy questions are real. The infrastructure decisions are real.

It’s not slowing down, and it’s important that it be done transparently. 

AI adoption inside government will be shaped as much by politics as by capability.

Technology moves quickly. Political incentives need to move with it.

Some light reading for when you start noticing the same theme showing up across sectors. 

If AI moves fast, leadership has to move faster - AI is already embedded in workflows, but leadership models are still built around control, approvals, and layered sign-off. Cheryl Cran argues that defining outcomes clearly and distributing decision authority is central to how organizations can adapt.

Canada summons OpenAI over failure to report mass shooter - After identifying a concerning ChatGPT account later linked to a mass shooting in British Columbia, OpenAI did not notify Canadian law enforcement. The federal AI minister called the decision “very disturbing” and requested a formal explanation of the company’s reporting thresholds. 

The woman fighting to reclaim her face from Albania’s ‘AI minister’ - Anila Bisha agreed to record content for a chatbot. Needless to say, she was a little surprised when Albania’s new AI government “minister” started giving speeches with her face and voice. Her legal fight pulls constitutional rights, consent, and digital identity into the spotlight. Once your face becomes data, does control become negotiable?

Alberta bets on scale to keep tech companies at home - Alberta’s tech ecosystem can produce startups with momentum. The challenge now is later-stage capital and ownership drifting outward. Provincial leaders are exploring co-investment strategies and government procurement reform to anchor companies locally as they grow.

The Insight Forum is Digital Journal's thought leadership platform, offering experts a dedicated space to share their perspectives with our audience across Canada, the U.S. and abroad. Members publish monthly articles showcasing industry insights and what they’re learning and seeing in their space.

Here is the mildly cynical but realistic part.

Governments act based on what they believe citizens will reward or punish at the ballot box. They adjust ambition, caution, and speed to political signals.

AI adoption inside government won’t be different. It will be shaped by what voters pay attention to, what they question, and what they let slide.

If AI in public systems feels abstract, it will stay a technical discussion among officials and vendors. If citizens treat it as central to service delivery, privacy, and national leverage, the incentives shift.

Oh, by the way. You’re the citizens. 

- David