Hi {{first_name}},

A data breach in Canada now costs an average of $7.11 million, the highest IBM has ever recorded here. That's the headline number (literally) from Jennifer Friesen's reporting on the new IBM report.

The other numbers that stuck out involve what AI can do to that cost. 

Companies using AI heavily across their security operations averaged $5.5 million per breach. The ones that weren't averaged $8.91 million. That's a $3.41 million gap on the same kind of incident, and the heavy users also detected and contained breaches weeks faster.

So while everyone's looking for the AI number that shows up as revenue, this is the other kind, the money you don't spend and the time you don't lose. It's a harder case to make to a board than a revenue line, but a $3.41 million gap is the kind of number a technology leader can use to build a case.

For more, including which sectors are paying the most and why supply-chain breaches cost the most of all:

→ Read: Average Canadian data breach cost hits record $7.11 million

The polite version of show me the money

Fidelity polled Canadian financial advisors during earnings season, and 42% of them said the thing they're watching most is whether AI spending is turning into actual revenue, ahead of guidance, capital spending and margins.

They're grading public companies, but it's the same question your board asks you once the AI budget gets big enough to notice. What did we get for it?

Most advisors are still optimistic. 84% called AI an attractive long-term theme and only 8% said it's overhyped. The worry is timing. 42% named stretched valuations the biggest near-term risk and 27% said adoption is coming in below what was promised.

For the full breakdown of what they're watching this earnings season:

→ Read: Financial advisors want AI spending to show up in revenue

EY Canada installed a quantum computer in one of its own buildings. The point isn't speed. It's that fraud detection and risk modelling for sensitive clients now runs on-site, inside Canadian borders, instead of in someone else's cloud.

As Jennifer Kervin reports, owning the machine gives EY control over where the data lives and who can touch it, which is what clients in banking, energy and government care about most.

Nobody's pretending the payoff is close. Raj Sharma, EY's global managing partner for growth and innovation, puts quantum's full potential five years out and says the work until then is getting the data and governance right. EY's running the systems on itself first before offering them to clients.

For more on what EY's building and who it's for:

→ Read: EY Canada grows its quantum capability to keep data in the country

Some light reading for when you're the only one on the status call

Chinese AI drives price competition among US labs OpenAI cut fees on its lightweight Luna model by 80% and Anthropic put out one near its top performance at half the price, with cheaper Chinese models like DeepSeek's V4 Flash pushing up from below. Analyst Jack Gold says the pressure is coming from enterprises that won't spend twice an employee's salary on the AI meant to help that employee. The models keep getting cheaper, and buyers have noticed.

Trump admin to review 'closed' AI models before release The White House is finalizing a security review for advanced AI models, and it only covers the closed ones from OpenAI, Anthropic and Google. The open models from Meta, Nvidia and Chinese labs like DeepSeek sit outside it, even though those are the ones anyone can download and run on their own hardware. That makes sense, right?

AI is disrupting Canada's legal sector, but lawyers are not disappearing 66% of Canadian law firms told Clio that AI has improved their revenue, and the work drying up is the repetitive, document-heavy kind that used to train junior associates and keep paralegals busy. This doesn’t mean chats with ChatGPT. Privacy rules under PIPEDA and Quebec's Law 25 are pushing firms toward secure, enterprise-grade tools rather than whatever's free.

Canada's vertical farming revolution: Can technology help feed a nation through climate change? Stacked-tray farms are turning up in Canadian warehouses and shipping containers, where growers use AI to run the grow rooms and robots to cover the labour they can't hire. Electricity is still the biggest cost, which is why a lot of well-funded vertical farms elsewhere have struggled to make the numbers work. For a country that imports most of its winter produce, growing lettuce close to the people eating it has real appeal.

The case for AI has mostly been a promise. Spend now, trust us, the value will show up later. The honest answer to "what's it worth" has been a shrug and a good story. As AI goes into production in more businesses, we're starting to get some numbers to work with.

A $3.41 million gap is something a technology leader can put in front of a board and defend.

— David Potter

If you have any questions, you can reply to this email. I read everything.

Do you have a story we should chase, or a tip you'd share with a peer?
Contact us here.

Know someone who should be reading this? Forward it on, or
they can sign up here.

This issue is presented by Calgary Economic Development.
If you want your brand in front of the people running technology across Canada, you can sponsor a future issue.