Jun 24, 2026
Think about all the people whose job titles contain the word ‘agent’: travel agents, insurance agents, real estate agents — the list goes on. But a growing number of homeowners are experimenting with agentic AI to manage the sale of their own homes, and some are saving enormous sums of money doing it. Could it work in Canada?
The $90K experiment that caught everyone’s attention
Stuart Thompson, a tech reporter for The New York Times, has become the most talked-about example of AI-assisted home selling. When he and his wife decided to list their Hudson Valley home, he skipped the traditional real estate agent and turned to Google’s Gemini AI chatbot instead.
Originally, the couple and several agents they consulted estimated their home would sell for roughly US$550,000 (C$755,000). Thompson wasn’t excited about handing over a 3% listing commission — as much as US$30,000 (C$41,200) — to a seller’s agent. So he began experimenting with AI.
Thanks for subscribing!
The best of Money.ca
delivered weekly
By signing up, you accept Money.ca Terms of Use, Subscription Agreement, and Privacy Policy.
Using Gemini, Thompson crafted his listing, sourced a photographer, staged the home and navigated the steps to post it to the Multiple Listing Service (MLS). Within 24 hours, showings were booked. Soon after, three offers arrived — all over asking. AI even helped him assess which to accept. The winning bid wasn’t the highest, but it was a solid offer above US$600,000 (C$823,000).
Thompson hired a lawyer to handle the paperwork. Even with that expense, he and his wife cleared US$90,000 (C$123,500) more than expected on the sale — and the buyers agreed to cover their own agent's 2% commission, saving another US$12,000 (C$16,500). In all, Thompson saved US$36,000 (C$49,400) in commissions.
Advertisement
“I’m persuaded that AI may well transform real estate agents into something more like travel agents,” Thompson wrote. “Once essential to navigating an opaque process, they could soon become more of a nice-to-have for busy people who want a more carefree experience.”
Robert Levine, CEO of strategic consulting firm ComOps, had a similar story. He used ChatGPT to sell his Florida home — and received five offers in five days, ultimately closing at nearly US$1 million (C$1.31M), about US$100,000 (C$137,200) more than agents had predicted.
Thinking of selling or refinancing before your move? Get personalized mortgage options from Homewise — they’ll find your best rate in minutes.
What this means for Canadians — and why commissions hurt more here
The commission math hits Canadian sellers especially hard. According to the Canadian Real Estate Association (CREA), the national average home price reached $702,079 in May 2026 — the highest monthly figure in two years.
At a typical 5% total commission (2.5% per side, as is common in Ontario), selling a home at the national average costs roughly C$35,100 in agent fees alone — before HST. Add Ontario’s 13% HST on that commission and the real cost rises to approximately $39,660. In the Greater Toronto Area, where the average home sold for C$1,069,700 in May 2026, a 5% commission with HST amounts to more than C$60,000.
In the Greater Vancouver Area, the commission structure is tiered: typically 7% on the first C$100,000 of the sale price and 2.5% on the balance. On a $900,000 Vancouver home, that works out to roughly C$27,000, or C$28,350 after factoring in GST. However, this figure is then split between the seller’s agent and the buyer’s agent — 3.875% on the first C$100,000 and 1.3375% on the balance for the former, and 3.125% on the first C$100,000 and 1.1625% on the balance for the latter.
“Commissions are often up for negotiation,” notes Rates.ca. “There’s also the option to find a brokerage that charges a flat fee, or you can sell the home yourself.”
Advertisement
Canada’s Competition Bureau, the federal agency that promotes fair market competition, has been investigating CREA’s commission rules since October 2024. It obtained a court order to examine whether the rules discourage buyers’ agents from offering lower rates or create conditions for ‘steering’ — the practice of directing buyers toward homes offering higher commissions. In February 2026, the Bureau expanded its investigation to include Greater Vancouver Realtors.
The Canadian catch: you still need a licensed agent or brokerage for MLS access
Here’s where Canadian sellers face a meaningful constraint that Thompson didn’t have to navigate in the same way: in Canada, private sellers cannot list directly on the MLS® System. Access to MLS® — the dominant database used by the vast majority of buyers and their agents — requires a licensed REALTOR® or a registered brokerage.
That said, flat-fee ‘mere posting’ services have become a popular workaround. For a one-time fee, often between C$299 and C$999, these brokerages will post your listing on the MLS® System and Realtor.ca on your behalf — without providing representation, staging advice or negotiation support. Services like For Sale By Owner Inc. (FSBO.ca), ComFree and One Flat Fee operate across most provinces.
This means a Canadian seller pairing AI guidance with a flat-fee MLS listing service could potentially replicate Thompson’s approach: use AI to craft a listing, research comparable sales on Realtor.ca, coordinate a photographer and navigate showings — then hire a real estate lawyer to handle the Agreement of Purchase and Sale (APS), as well as the closing.
In provinces like Ontario and B.C., hiring a real estate lawyer for the final paperwork is not optional — it’s legally required. Legal fees typically range from C$1,500 to C$3,000 for a standard residential transaction, according to RE/MAX Canada. Even at the higher end of that range, the math often still strongly favours the AI-plus-flat-fee approach.
Using AI to sell a home isn’t all smooth sailing
Thompson and Levine both acknowledged challenges using AI instead of a real estate agent.
There are some things a virtual assistant can’t do: show prospective buyers a home, touch up paint, move furniture or call buyers’ agents. All that work lands in the lap of a DIY seller.
In the comments on Thompson’s NYT article, one reader questioned how much time he actually saved. “You did all the legwork, calling, staging,” the reader wrote. “It sounds like a lot; how many hours did it add up to and how much would that time be worth at your normal rate of pay?”
Advertisement
Thompson also noted moments when he needed emotional support during the nerve-wracking sales process — a big part of what a real-life agent provides.
And AI agents can make mistakes. Thompson noted that Gemini initially gave him the wrong advice on paying the buyer’s agent a commission.
Ines Hegedus-Garcia, managing partner of the realty firm that represented the buyer in Levine’s sale, told Realtor.com that Levine could have made more with an agent. She argued that a professional would have negotiated a better deal during the bidding war, and that Levine’s ChatGPT queries were sometimes legally off-base. “’[The buyer’s agent] is like, no, that doesn’t apply in Florida,’” she said.
The same risk applies in Canada. Provincial real estate rules vary — what’s standard practice in Ontario under the Trust in Real Estate Services Act (TRESA) may differ from B.C. rules under the Real Estate Services Act, or from Quebec’s regime, which is governed by the Organisme d’autoréglementation du courtage immobilier du Québec (OACIQ). AI tools trained largely on U.S. data may not reliably reflect Canadian provincial differences.
Still, Levine has no regrets. “I’d recommend it to everyone,” he said. “ChatGPT is not coding. It is a conversation, and you’re going to have to have that conversation with a real estate professional if you want to go that direction anyway.”
What Canadian sellers should do before trying AI-assisted home selling
If you’re thinking about using AI to help sell your home, here are some steps to take first:
- Understand your legal obligations. In Ontario, using a real estate lawyer to close the transaction is mandatory. Review your provincial disclosure requirements — for instance, Ontario’s Seller Property Information Statement (SPIS), though technically optional, is frequently expected by buyers.
- Research your market with Canadian data. Use Realtor.ca (owned by CREA) to review comparable sales in your area. Unlike Zillow, which does not operate in Canada, Realtor.ca is the authoritative public-facing database for Canadian listings. For GTA sales history and data, HouseSigma is also a widely used tool.
- Price carefully. AI tools can help you research comparables, but they may not account for hyper-local factors that a licensed agent who knows your street might flag. Getting a formal appraisal (C$300 to C$500) or a Comparative Market Analysis (CMA) from a local agent before going it alone can add a useful sanity check.
- Use a flat-fee MLS service for exposure. Because private sellers cannot post directly to the MLS® in Canada, a flat-fee brokerage is typically the most cost-effective path to maximum buyer exposure. Research the service before committing: confirm what’s included, what buyer agent commission you’ll offer (agents often expect 2.5% to remain motivated to show your home), and what happens if you need to change your listing.
- Be ready for the emotional and logistical load. As Thompson and Levine discovered, selling without an agent means managing showings, fielding inquiries, negotiating offers and staying calm under pressure — all on your own timeline.
- Hire a real estate lawyer early. Don’t wait until you have an offer in hand. Engaging a lawyer before you list means you’re prepared to move quickly if a buyer comes in fast — which, in a hot market, they often do.
You May Also Like
- This 7-step plan from Dave Ramsey is designed to help you ditch debt, save more and build wealth — here’s how it works
- Prioritize these 4 critical investments and watch your net worth skyrocket
- Focus on these 3 ‘magic numbers’ to become a millionaire — and only on these numbers. How do you stack up?
- Millionaires under 43 are reshaping investing — just 25% of their portfolios are in stocks. Here’s where their money is going
The most expensive financial mistakes are often the ones you don't see coming. Join 19,000+ Canadians who get the money moves, risks and opportunities shaping their finances — delivered free each week. Subscribe now.
Laura Boast is an Associate Editor with Moneywise.com and a lifelong content creator who has reached international audiences at Discovery, CBC, Blue Ant Media, Bond Brand Loyalty and more.
