AI for Property Managers: 5 Jobs to Hand Over, and 3 to Keep (Ontario)
AI for property managers in Ontario: five office jobs to hand over this month, three that stay human, and the screening and notice rules that apply.
Property management is a job made of interruptions. A tenant texts about a dripping tap at 9 p.m. An owner wants to know why the roof quote went up. Three contractors send three quotes in three formats, and the move-out inspection for 4B still isn't written up.
Most of that is office work, and AI for property managers is good at office work this month, on the assistant you already pay for. It's also risky in a few specific places: tenant selection, legal notices, and how rents get set. This guide covers five jobs a small Ontario property manager can hand over, three that stay with a person, and the rules behind the line. We aren't lawyers and this isn't legal advice. Where a rule matters, we link the source.
AI for property managers: five jobs to hand over
None of these needs a tenant's name. That's why they made the list.
1. Maintenance replies, including the entry notice
Most maintenance messages need one of about twenty answers: we got it, here's who's coming, here's when, here's what to do until then. The Support Macro Library builds 25 reply templates from your policies and a couple of messages you've already sent. Swap its retail categories for yours (requests, entry, emergencies, move-in, move-out) and it writes the set in one pass.
Check the entry template line by line. Under the Residential Tenancies Act, a landlord entering for repairs gives at least 24 hours' written notice and enters between 8 a.m. and 8 p.m., and the Landlord and Tenant Board's interpretation guideline says the notice must state the date, the time and the reason. AI likes to write "between 9 and 3." Ontario's Divisional Court held that a six-hour window wasn't a specific time. Put a real time in the template.
2. Comparing contractor quotes
Snow removal, a roof, a boiler: the quotes never line up. The Vendor Comparison Matrix takes two to five quotes as pasted text and returns one table with the same rows for each (price, what's excluded, warranty, payment terms) plus the fees hiding in the fine print. The quotes are business documents, so nothing about a tenant goes in.
3. The monthly owner report
Owners want the same four things every month: what came in, what went out, what broke, what's coming. The Weekly Report Automator turns your scattered notes into that report in a fixed format. Change the period to monthly and the KPIs to rent collected, vacancies and open work orders. Paste unit numbers, not tenant names.
4. The move-in packet and tenant FAQ
Garbage day. Who to call at 2 a.m. when water comes through the ceiling. Whether tenant insurance is required. Say you manage 40 units across six buildings in Hamilton: the same dozen questions arrive every September. The FAQ and knowledge base builder turns a pasted list of them into a draft page. Write the answers from your lease and building rules, then check each one before it goes in a packet. A wrong answer about who pays for a repair becomes a dispute.
5. Turnover and emergency procedures
Between tenants there's an inspection, a deposit question, a cleaner, a painter, a lock change and a listing, and the order lives in one person's head. The SOP Writer asks six questions about the process and writes a numbered procedure someone else can follow. Do the after-hours emergency procedure second. It's the one a new hire needs on their first weekend.
The setup, before any tenant file goes near it
- Training off. In Claude on Free, Pro and Max: Settings > Privacy > Help Improve our AI models. In ChatGPT: Settings > Data controls > Improve the model for everyone. Both start switched on.
- Units, not people. "Tenant in 4B" works for almost every task above. If a job won't run without a name, a phone number or an income figure, it doesn't belong in a general chatbot.
- A written rule for staff. One page on which tools are allowed and what never goes in a prompt. Our AI policy template is written for a Canadian office.
- One workspace per job. A Claude Project for maintenance replies, one for owner reports. The lesson on how to use Claude Projects covers what to upload.
Three jobs that stay with a person
Choosing a tenant
Ontario's Human Rights Code lets a landlord use rental history, credit references, credit checks and income information. Regulation 290/98 adds a condition: income information only alongside the other three, and weighed together with them. The Ontario Human Rights Commission's background paper on minimum income criteria walks through the decisions that found rent-to-income ratios and minimum income cut-offs discriminatory. A tool that ranks applicants on "rent is under 30% of income" is applying that rule, faster and with less scrutiny.
Privacy points the same way. The federal privacy commissioner's tips for the rental sector say not to ask for a SIN unless the law requires it (mark it optional if you do) and to stay away from social media background checks. A chatbot can tidy your written screening criteria. The yes or no on an applicant stays with you.
Anything the Board has a form for
An N4 for unpaid rent, an N1 for a rent increase, an N12: use the Board's forms and fill them in yourself. General chatbots blend Ontario with British Columbia and a dozen US states, and they're often a year behind. For 2026 the rent increase guideline is 2.1% for rent-controlled units, with at least 90 days of written notice. Ask a chatbot and you may get last year's number. AI can write the polite cover note that goes with the form. The form is yours.
Setting the rent
The Competition Bureau spent much of 2025 looking at revenue management software from RealPage and Yardi. On November 10, 2025 it closed the case: the tools hadn't been used widely enough to substantially harm competition, and landlords had cut back on them since late 2024. It stayed concerned and published guidance for landlords and property managers. The practical version for a small manager: set rents from comparables you gathered yourself, treat any software recommendation as advice you can override, and don't feed your non-public rents and vacancies into a tool that pools them with competitors'.
Where this breaks
AI drafts fast and checks nothing. It doesn't know your lease, your building rules or this year's guideline unless you paste them in, and it will produce a confident wrong answer about who pays for a broken fridge. Every message that touches a tenant's rights gets read by a person before it's sent.
This guide is for residential rentals in Ontario. Condo management and commercial leases work under different rules, and a phone bot that answers your office line is a separate purchase with its own costs; our AI receptionist cost guide covers that one.
Where to start
Build the maintenance reply set first, with a real time in the entry notice. It saves the most typing in week one. For a scored view of where AI fits across the rest of the business, the free AI Opportunity Audit covers seven areas in one sitting.
Common questions
- Can property managers use AI to screen tenants in Ontario?
- As a filing clerk, yes. As the decision-maker, it is risky. Ontario's Human Rights Code lets landlords use rental history, credit references, credit checks and income information, but Regulation 290/98 allows income information only alongside the other three, and Ontario human rights decisions have found rent-to-income ratios and minimum income cut-offs discriminatory. A scoring tool that ranks applicants on an income ratio applies the same rule, faster. Keep the selection decision with a person and apply the same written criteria to every applicant.
- Can I put tenant information into ChatGPT or Claude?
- Not as the tools ship. The federal privacy commissioner's tips for the rental sector say to collect only what you need, protect it, and use it only for purposes you told the tenant about. Pasting a rental application into a consumer chatbot is a new use the tenant never agreed to. Strip names, unit numbers and contact details first, and switch off model training: Help Improve our AI models under Settings > Privacy in Claude, Improve the model for everyone under Settings > Data controls in ChatGPT.
- Can AI write an N4 or a rent increase notice?
- Use the Landlord and Tenant Board's own forms for anything the Residential Tenancies Act requires, and fill them in yourself. A chatbot often mixes Ontario rules with other provinces or US states. For 2026 the rent increase guideline is 2.1% for rent-controlled units, and the tenant needs at least 90 days of written notice. AI is fine for the friendly covering letter that goes with the form.
- Is it legal to use AI to set rents in Canada?
- Yes, with care about data. On November 10, 2025 the Competition Bureau closed its investigation into RealPage and Yardi pricing tools, finding they had not been used widely enough to harm competition, and published guidance for landlords and property managers. The core of it: treat recommendations as advice you can override, and keep non-public rent and vacancy data from flowing to competitors through a shared tool.
- What are the best AI tools for property managers?
- For a small portfolio, the assistant you already pay for (Claude, ChatGPT or Microsoft 365 Copilot) handles the office work in this guide: maintenance replies, vendor quotes, owner reports, tenant FAQs and procedures. Check what your property management software already includes before you buy a separate AI product, and keep tenant data inside tools whose terms you have read.
More from the blog
Want this customized and automated for your business?
We take the tools in this toolbox and wire them into your business: your data, your brand voice, running on autopilot.
Talk to Vexlo