The Toronto edition: the birthplace of deep learning itself still can't get its own singles to move faster
Every city in this series has had some claim to AI credibility — one hosts it, one funds it, one manufactures its physical chips. Toronto's claim predates all of them. This is where Geoffrey Hinton did the foundational deep learning research at the University of Toronto that underpins essentially every modern AI system, work for which he won the 2024 Nobel Prize in Physics. The federal government partnered with U of T and Hinton himself to launch the Vector Institute in 2017, now widely described as the heartbeat of Canadian AI research, and Ottawa's 2025 budget committed close to $1 billion toward sovereign AI infrastructure specifically to keep building on that foundation. The Toronto–Waterloo corridor is now the third-largest technology cluster in North America, home to more than 2,000 startups and pulling in 40 to 60% of all Canadian venture capital. If any city on Earth has earned the right to say it didn't just adopt AI, it invented the version everyone else is now building on top of, it's this one.
Which makes it the most fundamental version yet of the question this series keeps asking: if the city where deep learning itself was born hasn't sped up its own dating culture, what does that tell you about where the technology's real limits sit?
Because Toronto's dating scene has a specific, well-documented character, and speed isn't part of it. Local matchmaking guides describe the culture bluntly: polite to a fault, app-heavy, and — this is the part that surprises people — quietly favorable for men, running around 1.04 women per man in the 18-to-34 bracket, a genuinely rare tilt among major North American cities. But Statistics Canada data tells a more layered story once you widen past that single bracket: across all ages, Toronto actually has more single women than single men citywide, with the balance flipping in men's favor only in the younger cohorts and reversing sharply after 40. That's two different, genuinely accurate statistics describing two different dating markets depending on which decade someone's in — the kind of layered, cohort-specific nuance a matching algorithm trained on a single citywide average tends to flatten out entirely.
What Toronto's dating culture actually struggles with isn't scarcity. It's momentum. As Luvo Matchmaking's own read on the city puts it, the issue in Toronto usually "isn't a lack of compatibility. It's a lack of momentum" — a city where people are thoughtful and cautious by default, where conversations can drift pleasantly for weeks without anyone proposing an actual plan. That's a real, specific cultural pattern, distinct from any gender-ratio math, and it's exactly the kind of texture an algorithm optimized for compatibility scoring has no mechanism to fix, because the problem was never who to match — it's getting two well-matched, equally polite people to actually commit to a Tuesday.
Now compare that to what Toronto-adjacent AI is genuinely excellent at, which is the same logistics category this series keeps finding everywhere else. Nationally in the U.S., Phocuswright found 56% of leisure travelers used AI to plan at least one trip this year, and 78% of people who tried it booked something based primarily on its recommendation — a clean, structured, bounded problem, built on exactly the kind of research infrastructure Toronto pioneered. Pearson sits inside a metro that quite literally trained the models solving that category of problem for the rest of the world.
Matchmaking has never been that kind of problem, and the industry's clearest cautionary tale doesn't need a Toronto-specific example to land. Justin McLeod, who built Hinge into the most dominant dating app in North America, left to launch an AI-first matchmaking startup backed in part by Match Group — and has said plainly that even his new venture treats AI as a coach, not a decision-maker. Three Day Rule, fifteen years into human-run matchmaking before adding an AI-matching layer, drew early criticism for a specific failure: both sides of a match receiving identical AI-generated opening lines, meaning two people's software did the talking while the actual humans, already prone to letting conversations drift, sat it out entirely. In a city whose default setting is already cautious and slow-moving, that's the last thing worth automating.
None of this argues AI has no place in Toronto dating — coaching tools that sharpen a profile or catch a self-defeating pattern remain a legitimate, evidence-backed use case, distinct from trying to automate the actual push forward. But Toronto might be this series' most fitting final word on where that line sits: the city that trained the deep learning models running underneath most of the AI industry, and still one where, per Luvo's own read, the goal "isn't more introductions. It's introductions that move forward" — something no model, however foundational, can supply on its own. That's the case for Relish's structured evenings and Luvo's matchmaking practice here in Toronto, sourcing introductions from real conversation and observed interactions rather than a database that can optimize a match but can't make anyone actually text back. Toronto will keep training the AI that runs the rest of the world. Getting two thoughtful people to stop being cautious with each other is still, even in the birthplace of deep learning, a human job.