Canada
30 projects in 4 cities
The atlas documents 30 projects across 4 cities in Canada, the earliest dating to 1985: 3 with a documented success, 1 documented failure, 2 scaled back, 24 still running without a verdict. Every entry carries an evidence grade — the strongest here is peer-reviewed (2 entries) — and links its sources.
What the record shows
Ask who holds the lever before asking whether the technology works — in the Canadian record they are rarely the same government. Toronto's fare card is built and operated by Metrolinx, an agency of the Government of Ontario: the TTC buys it as a managed service, the city's own Auditor General estimated at least $64M CAD in passenger revenue lost in 2018, and the agreement's base term can simply expire in 2027. Montreal's inversion is starker still: photo radar has run on its streets since 2009 and the city owns none of it — Quebec picks the sites, and provincial guidance states that municipalities cannot apply for a device. The mayor asked for 300 cameras in 2024 and got no power to install one. The same altitude can also do what no city could: the ARTM, the regional authority that has owned Greater Montreal's ticketing since 2017, collapsed 17 fare grids and more than 750 fare products into four zones.
The sharpest entry here is a programme that was working when it was switched off. Toronto's speed cameras have the rarest grade in this record: a peer-reviewed evaluation across 204 locations, by SickKids and Toronto Metropolitan University, found the share of vehicles speeding down 45%. What ended it was a saw and a statute, not evidence: vandals felled 16 cameras in a single night in September 2025, and Ontario's Bill 56 repealed the enabling law, stopping all ticketing on 14 November 2025. By June 2026 the city measured speeds up at 101 of 104 sites. Neither the province nor the vandals disputed the measurements. The pattern repeats wherever enforcement is the moving part: on King Street, a 2023 University of Toronto study counted about 6,800 illegal turns a day with under 0.3 percent ticketed, the gains eroding as ticketing lapsed. Vancouver wrote binding carbon-intensity caps for its large buildings — then a new council majority voted 7–4 in May 2026 to pause enforcement before a single building had ever been fined.
Canada holds both ends of the driverless-metro story. Vancouver's SkyTrain has run without drivers since it opened in December 1985 — none of the staffed transition other automated metros needed. At the other end, Montreal's REM is a driverless light metro belonging to CDPQ Infra, the Quebec pension fund's infrastructure arm, paid by the regional transit authority for every passenger-kilometre travelled; Quebec's Auditor General reported in November 2024 that the provincial contribution will supply more than half of the ridership revenue the REM collects over its operating life. Between the two sits Toronto's retrofit, which shows what automation usually buys an existing system: Line 1's moving-block resignalling finished in September 2022, seven years past its target, with an operator still in the cab — what automation actually removed was the second crew member.
Who owns what is never one question here — the hardware, the data and the fare lever can each land somewhere different. Both big bike-share systems trace to the same collapse, the Montreal-built equipment's maker going bankrupt in January 2014 owing about CAD 50M. Montreal bought the local assets for CAD 11.9M and created a non-profit to run them; Toronto, owed most of a CAD 4.5M loan, took its system over outright. Today Toronto's parking authority owns the bikes and docks and sets the fares — community-housing tenants ride a year for CAD 5 against the standard CAD 105 — and the city publishes the trip data itself under its own licence, while Montreal still buys BIXI's hardware yet the copyright in the published trip history is claimed by the operator. The same fine print runs through the open-data catalogues: Montreal names a licence on every one of its 405 datasets, while Toronto's bespoke municipal deed reached 115 of 554 records.
This record is unusually legible because its auditors keep writing it. Ontario's Auditor General reported in 2014 that the province's full-population smart-meter rollout had cost close to CAD 2 billion against an initial CAD 1 billion projection, on a business case written after installation began; the honest payoff, from a peer-reviewed study, was on-peak demand down 2.6%. Montreal's auditor judged the city's CA$356M water-metering contract "trop vite, trop gros et trop cher", the city cancelled it in September 2009, then bought roughly the same meters for about CA$55M over the following decade — while the network's losses sit at 30%, where they were in 2015. The audits also mark where the record goes dark: no audit has tested whether a Toronto 311 request marked closed was actually fixed, and Montreal's GPS-tracked snow operation warns drivers before a parking ban while the orange street sign, not the feed, decides who is towed — CBC counted close to 3,000 vehicles towed outside the hours the app had indicated between 2015 and 2018.
The country's most famous smart-city decision was a refusal. Alphabet's Sidewalk Labs planned one of the world's most heavily instrumented neighbourhoods on Toronto's waterfront, a contested "urban data trust" included — and abandoned it in May 2020, officially over pandemic economics, after years of privacy backlash. What it left behind is visible: Toronto's council adopted a digital-infrastructure framework — six principles the CTO said the failed waterfront project catalysed — whose one visible trace so far is an alignment table on a single procurement. Vancouver's caution reads the same way: its 2026 Google traffic-signal pilot is advisory only, the city shared none of its own signal-timing data, and it can walk away on 60 days' notice. Ottawa's federal experiment, a Smart Cities Challenge launched in November 2017, paid out $75 million in prizes in May 2019, $50 million of it to Montréal; no entry here yet tracks what that prize built. The baseline layer is still going in: Winnipeg switched on fleet-wide live bus tracking in March 2026. If the entries below agree on one lesson, it is that in Canada the instrument is rarely the contested part: what decides how the story ends is which government holds the lever, and how long it chooses to keep holding it.
Sources beyond the atlas
Facts above that do not come from a linked entry rest on these:
- Government of Canada release, 14 May 2019 — Smart Cities Challenge winners — challenge launched November 2017, more than 200 communities responded, 20 finalists; $75 million in prizes, the largest $50 million to Montréal for mobility and access to food
The shape of Canada
Computed from the 30 entries below — a profile, not a score.
Cities
All 30 projects
✔ Every published entry is checked against the public sources it cites before publication.
Show 18 more projects in CanadaShow fewer
Deeper reads
Where Canada appears in the atlas's own analysis.
📎 Cite this country record
1001 Smart Cities (2026). “Smart-city projects in Canada.” The Smart City Atlas. https://1001smartcities.org/countries/canada/ (last verified 2026-08-27). Data: CC BY 4.0.
The underlying data is free to reuse with attribution — see the open data page.
