Last week, we did something you almost never see in the fight against voice fraud - we got everyone in the same room.
Rogers Communications and Hiya hosted a half-day executive forum in Toronto on a problem that's getting harder by the month, voice fraud in Canada. We were joined by mobile network operators, financial institutions, security partners, and regulators. The people who too often have to work this problem from their own corner, finally getting the opportunity to discuss, question, and collaborate, together.
The day started with the hard truth about how much bank fraud has increased in Canada, told firsthand by the banks living it. But here's what made this forum different: we didn't stop at the problem. Rogers Bank stood up and showed the solution, how it's using call authentication to prove a call really is coming from who it says it is.
That matters. It means call authentication isn't theoretical anymore. It's real, it's running today, and Canada is set up to build the trusted-voice standard the rest of North America can follow.
Phone fraud in Canada
Hiya Product Manager Jonathan Nelson presents data on phone fraud in Canada at the Rogers × Hiya Voice Fraud Forum in Toronto.
Our Hiya Product Manager Jonathan Nelson opened with the numbers, drawn from Hiya's State of the Call 2026 report, a survey of more than 12,000 consumers across six countries, Canada included. His message was blunt: voice fraud in Canada has hit a tipping point, and the data says it's still accelerating. A few proof points:
-
Canada is getting hit hard. Canadians told us they get about 5 unwanted calls a week. One in ten lost money to a phone scam in the past year, losing CA$1,200 on average. And the dominant scam has shifted, from government impersonators, to Amazon, and now to your bank.
-
AI is pouring fuel on the fire. Nearly a third of Canadian consumers (31%) say they got a deepfake voice call in the past year. Their verdict on who's winning right now? The scammers, not the operators.
- Trust in the phone call is draining away. 88% of Canadian consumers won't answer a call from a number they don't recognize. That's a real problem for every bank and business that needs to reach its customers by phone.
The fraud landscape, as the banks see it
A panel brought key players together, including Rogers Communications, TD Bank, BMO, and credit scoring agency FICO.
Next came a panel of people on the front lines, Canadian financial institutions fighting voice fraud every day. A few things stuck with us:
-
Fraud starts upstream, but banks foot the bill. Criminals aren't breaking into the bank. They're going straight at the bank's customers over the phone, which sidesteps a lot of the bank's defenses. The fraud may start upstream, but the bank still absorbs a problem it can't fully control: reimbursements, investigations, swelling contact-center volumes, and the brand damage when a customer's trust is gone for good.
-
AI changed the game in two short years. It's made scams faster and far more convincing. One of our panelists described fielding more than 100 calls from an AI bot impersonating its own fraud team, adjusting its answers in real time. As AI gets better at mimicking customers, older defenses like voice biometrics and security questions get weaker.
- Coordination is the answer. Banks want consistency, so a customer knows what a real bank call looks like no matter who's calling. They want real-time signal sharing, knowing a line's been forwarded, or that a customer is on a live call right now. The gap that needs to be addressed is how can we make this work across every operator?
How Rogers Bank is authenticating its calls
In session three, Rogers Bank CTO Sougat Ray walked through how the bank is actually doing this. The short version:
-
Before placing a call, the bank sends an encrypted "pre-announcement" to Hiya, which is an indication to Hiya that Rogers Bank is about to place a call from “A” number to “B” number.
-
On the receiving end, the call gets matched against it.
-
In this first phase, a fraudster can still spoof the number, but can't fake the pre-announcement. So authenticated calls show branding, while spoofed calls get their branding stripped. In the next phase, spoofed calls won't ring at all. They'll be blocked at the network.
None of this was friction-free, and Sougat didn't pretend otherwise. Branding has to be supported by both the operator and the device.
The takeaway for the room was simple. This isn't a slide. It's live. It's what coordinated action between a bank and an operator actually looks like, and it's a playbook other Canadian banks and operators can run.
Fighting fraud together
We closed with one representative from each side of the problem: TD (bank), Rogers (network), and SecureLogix (security partner). What they landed on:
-
A coordinated defense is the only one that works at scale. Every player in the call flow, the operator, the bank, and everyone in between, has to be in. Go it alone and AI-powered fraud keeps gaining ground. Get everyone in, and authentication stops being a project and starts being the expectation.
-
STIR/SHAKEN won't carry the near term. Legacy landlines, coverage gaps, and multi-hop international calls mean Canada's STIR/SHAKEN rules can't currently be leaned on to stop fraud calls. The practical path right now: register your legitimate numbers, add out-of-band authentication, and get consistent branding across operators.
-
Try things that work and share what you learn. Panelists suggested running a pilot project where you set a baseline for spam-tag rates, block rates, and pickup rates, and then turn on branding for some numbers and measure the lift. Tools such as Hiya's Cross-Carrier Call Inspector will show how phone numbers actually display to call recipients.
The message from the day was unanimous: fraud is a shared problem, so it needs a shared defense. And the technology to authenticate calls and bring trust back to the voice channel? It's already here.
Learn more about call authentication and Hiya’s Branded Call →
