AI for real estate agencies in Australia: what actually works

Short answer: AI earns its keep in five places inside an Australian real estate agency: drafting listing copy and campaign material, triaging portal enquiries, preparing appraisal comparables, generating vendor reports, and handling property management arrears and maintenance. Everything else is either immature, a compliance liability, or both.

Listing copy and campaign drafting

This is where most agencies start, because the return is immediate and the risk is low. A general-purpose AI assistant on a business tier (Claude for Business, ChatGPT Team, Microsoft Copilot for Microsoft 365) can produce a serviceable first draft of a listing description, social media post or email campaign in under a minute. The agent reviews, adjusts anything property-specific, and publishes under their name.

The return is not in the quality of the first draft. It is in the 20 to 40 minutes recovered per listing that used to go into staring at a blank page. Multiply that across a team listing 10 properties a week and the time adds up fast.

Purpose-built tools exist too. BoxBrownie offers 10 free AI listing descriptions per month in 150, 200 or 250-word lengths, generated from a submission form with the property's features.¹ Rex Software's AI suite, launched in mid-2026, generates ad copy, SMS and email content from listing data in one click, embedded directly in the CRM.² AiRealEstate.Tech, built by an Australian agent, generates headlines, descriptions, social captions and email subject lines from $49 a month.³

What it costs to set up: a business-tier AI subscription is roughly $30 to $50 per user per month. The real setup cost is writing the prompt templates that produce output in your agency's tone with the right disclaimers and property-specific detail. A principal or senior agent needs to spend half a day building those templates, then share them with the team. Without them, every agent will produce inconsistent output and the quality control problem moves from writing to editing.

One hard rule: never paste buyer or vendor personal information into a free-tier AI tool. Free tiers typically retain input data for model training. Business and enterprise tiers generally do not, but read the data processing terms. This is a Privacy Act obligation, not a suggestion.

Enquiry triage and first response on portal leads

A listing on realestate.com.au or Domain generates enquiries around the clock. The agencies that convert the most are the ones that respond within minutes, not hours. AI can close that gap without adding headcount.

RiTA, now owned by Cotality (formerly CoreLogic), connects to your CRM and uses AI to score your contact database, generate prioritised daily call lists, and run personalised two-way SMS conversations at scale. It integrates with Rex, Agentbox, VaultRE, Box+Dice, Eagle and Inhabit.⁴ ActivePipe takes a different approach: behaviour-triggered email nurture with an intent engine that surfaces buying and selling signals from engagement patterns.⁵ Rex Software's AI Prospecting module, launched mid-2026, analyses CRM data to identify which contacts to call, why, and what to say.⁶

What it costs to set up: pricing for RiTA and ActivePipe is not publicly listed and requires a direct quote. Rex AI is bundled into the Professional CRM tier. The real setup cost is the same as any automation: mapping which enquiry types get an instant response, which get a human callback, and which go into a nurture sequence. Budget a day for a sales manager to define those rules, then a week of running both the old and new processes in parallel before switching over. If phone answering is the specific gap, rather than email and SMS, the AI receptionist buyer's guide covers the dedicated products and their pricing.

Appraisal and comparable-sales prep

Preparing a market appraisal used to take an experienced agent 90 minutes of pulling comparable sales data, formatting a CMA report, and writing the market commentary. AI and automated valuation models have cut that to under 15 minutes.

Cotality RP Data (formerly CoreLogic) is the dominant platform, with 10 million property records, IntelliVal automated valuations and branded CMA reports from $180 to $320 a month depending on the tier.⁷ PropTrack, owned by REA Group, powers the property valuations on realestate.com.au using an AVM trained on 12.6 million residential properties.⁸ Pricefinder, now part of the Domain/CoStar ecosystem, offers branded CMA reports and comparable sales search from roughly $175 plus GST a month.⁹

Realtair Pitch is not an AI tool, but it streamlines the presentation side: a customisable appraisal builder that pulls market data from CoreLogic, PropTrack and Pricefinder into a branded digital proposal, from $189 a month.¹⁰ For the market commentary itself, a general-purpose AI assistant can draft the narrative around the data, then the agent adds the local knowledge that no model has.

What it costs to set up: most agencies already subscribe to at least one data platform. The gap is usually in how it is used. If your team is manually downloading PDFs and copy-pasting comparable sales into a Word document, the setup cost is a day of someone building the CMA templates within the platform and training the team to use them.

Vendor reporting and campaign updates

The weekly vendor report is one of those tasks that takes 20 minutes per property but nobody bills for. It is also one of the easiest to accelerate with AI, because the inputs (campaign reach, portal views, enquiry count, open home attendance, offers received) are structured data that a language model can turn into clear, client-facing language in seconds.

Rex Software's Rex Hub gives vendors a branded, self-service portal with real-time campaign data, and agents can generate a vendor report with one click.¹¹ Campaigntrack, used by more than 4,000 offices across Australia and New Zealand, handles multi-channel marketing coordination with vendor-paid advertising tracking and automated PDF reports.¹²

For agencies without those platforms, a general-purpose AI assistant does the job. Paste in the campaign numbers, ask for a two-paragraph vendor update in your tone, review it, and send. The agent adds the market context ("three of the four groups at Saturday's open home were downsizers from the same suburb") that the data cannot provide.

What it costs to set up: if you already have a CRM that tracks campaign data, the setup is a prompt template and a 15-minute training session. If your campaign data lives in spreadsheets, the first step is consolidating it into one place, which is a workflow problem, not an AI one.

Property management arrears and maintenance triage

Property management is where AI is moving fastest in Australian real estate, because the workflows are high-volume, rules-based, and the cost of slow response is measurable in arrears and lost tenants.

PropertyMe's AiMe suite is the most widely adopted. Reply with AiMe drafts context-aware email responses matched to the property manager's tone. Bills AI extracts amounts, dates and descriptions from supplier invoices, including handwritten tradesperson notes. AiMe Comply, launched in February 2026, answers tenancy law questions with state-specific legislative references for NSW, VIC, QLD, SA, WA, TAS and the ACT. The AI features are in the Advanced tier at $215 a month per 100 properties.¹³

Kolmeo takes a different approach by including all AI features in every plan. Labs Chat is a tenant-facing AI that answers rental questions, triages maintenance requests with photo documentation and urgency assessment, and operates after hours. Labs Check monitors legislation changes across all Australian jurisdictions in real time. Kolmeo Sentinel, launched April 2026, is a standalone compliance monitoring tool available for $99 a month or free with a Kolmeo subscription. The core platform starts at $1.20 per property per month.¹⁴

What it costs to set up: if you are already on PropertyMe or Kolmeo, the AI features are a plan upgrade and a configuration session. If you are migrating from another platform, the setup cost is the migration itself: trust accounting data, tenancy records, owner and tenant contacts. Budget two to four weeks for a migration of 200 to 500 properties, depending on data quality.

Where it fails and what to never automate

Every vendor article about AI in real estate ends with "the future." This is not that section. This is the list of things that do not work today and that an agency should not attempt to automate, regardless of what a sales demo implies.

Price guidance and underquoting exposure

Automated valuation models are useful for preparing a comparable-sales analysis. They are not a substitute for an agent's professional estimate of selling price. In NSW, the Property and Stock Agents Act 2002 requires the agent's estimated selling price in the agency agreement to be "reasonable," with underquoting penalties of up to $110,000 or three times the agent's commission, whichever is higher, as of June 2026¹⁵. In Victoria, the Estate Agents Act 1980 requires a Statement of Information with comparable sales and a range of no more than 10%, with penalties exceeding $48,000¹⁶. In Queensland, section 212 of the Property Occupations Act 2014 prohibits false or misleading representations about property with a maximum penalty of 540 penalty units¹⁷.

None of these Acts mentions AI. None provides a defence for having relied on an automated valuation. The estimate must be the agent's own professional judgement, and the agent is personally liable if it turns out to be unreasonable. Use AVM data to inform your estimate, not to replace it.

Digitally altered listing images

AI image tools can remove power lines, add trees, brighten rooms, and replace furniture. Using them to misrepresent a property's actual condition is misleading conduct under section 18 of the Australian Consumer Law,¹⁸ with penalties of up to $1.1 million for companies. Section 30 of the ACL specifically covers false representations about land.

NSW Fair Trading's advertising guidelines already prohibit agents from digitally removing or adding features to change a property's appearance¹⁹. A separate bill targeting AI-altered images in rental listings was introduced to NSW Parliament in July 2025 but has not yet passed. In Queensland, the Office of Fair Trading sent a direct communication to every real estate licensee in June 2026 announcing targeted compliance checks from August 2026, with a specific focus on false or misleading representations arising from AI and digital enhancements in property marketing. The QLD OFT published a best practice guide addressing AI-generated content in listings.

Virtual staging is widely accepted, provided it is clearly disclosed. The industry standard is a watermark or caption ("virtually staged for illustrative purposes"). The line is between showing what a room could look like (acceptable, with disclosure) and misrepresenting what it does look like (not acceptable, with or without disclosure).

Neither realestate.com.au nor Domain has published a standalone policy specifically on AI-generated listing imagery as of September 2026. Both portals operate under the general ACL framework and state Fair Trading requirements. The Treasury's October 2025 review of AI under the Australian Consumer Law²⁰ confirmed that AI-generated content, including images, falls under existing consumer protections.

Tenant selection decisions

AI-powered tenant screening tools are growing in Australia. An AHURI study published in January 2026 identified 57 rental PropTech products operating in the Australian market, many using automated decision-making to rank applicants²¹. The scoring criteria and weighting are rarely disclosed to applicants.

The problem is not the technology. It is that algorithms trained on historical tenancy data can perpetuate existing bias. Refusing a tenant based on a protected characteristic, whether the refusal comes from a person or a model, violates anti-discrimination legislation in every state: the Anti-Discrimination Act 1977 (NSW), the Equal Opportunity Act 2010 (VIC), the Anti-Discrimination Act 1991 (QLD), and their equivalents in other jurisdictions. The agency is liable regardless of whether the discrimination was human or automated.

From 10 December 2026, the automated decision-making transparency provisions of the Privacy Act²² will require agencies to disclose in their privacy policies how personal information is used in decisions made by or substantially supported by computer programs. If your agency uses AI to triage or rank tenant applications, your privacy policy will need to say so.

Anything that goes out under the agent's licence unread

This is the meta-risk behind all the others. An AI tool can draft a listing description with a fabricated school in the catchment. It can generate a market commentary with an invented sales figure. It can produce a vendor letter with advice the agent's licence does not authorise them to give. The REIQ has cited a real incident where AI created non-existent schools in a listing description²³.

If it goes out under your name or your agency's brand, you own it. AI is a drafting tool. The review step is not optional.

The compliance reality for Australian agencies

Most articles about AI for real estate treat compliance as a paragraph at the end. For an Australian agency, it is the section that matters most, because the regulatory environment is specific and enforceable.

Privacy Act obligations on portal lead data

When a buyer enquires through realestate.com.au or Domain, the agent receives the enquirer's name, email, phone number and enquiry message. Entering that data into a third-party AI tool is a disclosure of personal information under Australian Privacy Principle 6 of the Privacy Act 1988.²⁴ The OAIC's guidance on commercially available AI products is explicit: entering personal information into a publicly available AI chatbot constitutes disclosure, because the information becomes accessible to others outside the organisation²⁵.

If the AI tool's servers are overseas (ChatGPT and Claude both process data in the United States), cross-border disclosure obligations under APP 8 apply. The agency remains legally responsible for how the information is handled after transfer.

From 1 July 2026, real estate agents became reporting entities under the AML/CTF Act Tranche 2 reforms. Under section 6E of the Privacy Act, reporting entities lose the small business exemption regardless of turnover. This means that effectively all real estate agencies in Australia are now covered by the Privacy Act for their AML/CTF identity verification activities. For agencies above $3 million turnover, the Act applies to all personal information handling, including portal leads.

If you have not started on an AI policy, that is the first step.

State agent conduct rules

The Property and Stock Agents Act 2002 (NSW), the Estate Agents Act 1980 (VIC) and the Property Occupations Act 2014 (QLD) all prohibit false or misleading representations about property. None of them mentions AI. All of them apply to AI-generated content, because the obligation sits with the licensed agent, not the tool. A listing description generated by AI that misrepresents the property's features, location or condition exposes the agent to the same penalties as if they had written it themselves.

The practical implication: every piece of AI-generated content that goes out under the agency's name needs a human review step. Listing descriptions need checking against the actual inspection notes. Market commentary needs its sales figures and suburb statistics verified. Vendor reports need the campaign data confirmed before reaching the client.

AI-generated imagery and Fair Trading enforcement

Queensland is leading enforcement. The Office of Fair Trading's June 2026 communication to all licensees announced targeted compliance checks from August 2026 focused specifically on AI and digital enhancements in property marketing. The accompanying best practice guide prohibits using AI to remove unattractive features, add non-existent elements, or alter images to misrepresent a property's condition.

NSW Fair Trading's existing advertising guidelines already prohibit digital modification that changes a property's appearance. A separate bill addressing AI-altered images in rental listings was introduced to Parliament in July 2025 but has not yet been enacted.

Victoria's Consumer Affairs has not issued AI-specific guidance for estate agents as of September 2026.

No Australian state has passed AI-specific real estate legislation. The enforcement is through existing consumer protection frameworks (the ACL at the federal level and the state agency Acts) applied to new technology. The REIQ's position is clear: "AI should be seen as a drafting and support tool only, not a substitute for professional judgement or due diligence."

How to run a two-week trial on one workflow

If you have read this far and want to test one AI workflow before committing to anything, here is the two-week trial we run with agencies at Bulletproof, and it works for an agency of any size.

Week 1: pick one workflow and measure it.

• Choose one of the five categories above. Listing copy is the safest starting point because the compliance risk is lowest (no personal data enters the tool if you are describing a property, not a person) and the time saving is immediately measurable.
• Time the current process. How long does it take your team to write a listing description, social post and email campaign for a new listing? Write that number down.
• Set up the tool. A business-tier AI subscription takes 10 minutes. The prompt templates take longer: build three (listing description, social caption, email blast) and test them against two or three current listings.
• Run it alongside the manual process for one week. Do not switch over. Run both and compare.

Week 2: compare and decide.

• Time the new process over the same volume. How long does the AI-draft-plus-review workflow take versus writing from scratch?
• Check quality. Read the AI-assisted listings alongside the manually written ones. Are they on brand? Did any property detail need correcting? How much editing did the agent actually do?
• Calculate the return. If the tool saves each agent 30 minutes per listing, and your team lists eight properties a week, that is four hours recovered. At an average agent cost of $70 an hour, that is $280 a week, or about $14,500 a year, for a tool that costs $50 a month.
• If the numbers work, switch over. If they do not, try a different workflow.

The point is to start with the boring, repetitive work rather than the ambitious idea. The agencies that get value from AI early are not the ones chasing the newest features. They are the ones that automated the task everyone hated, measured the result, and moved on to the next one.

Frequently asked questions

Can I use AI to write listing descriptions in Australia?

Yes, and many agents already do. The obligation is accuracy: every AI-generated description must be reviewed against the actual property before publication. Section 52 of the NSW Property and Stock Agents Act, section 212 of the QLD Property Occupations Act and equivalent provisions in other states prohibit misleading representations, whether written by a person or generated by a tool. Use AI to draft, then check every detail.

Will AI replace real estate agents in Australia?

No. AI automates drafting, data handling and first-response communication. It does not replace the professional judgement required for pricing, negotiation, vendor management or the client relationships that win listings. The agencies that use AI well will do more high-value work with the time they recover from administrative processing, not fewer people.

What AI tools do Australian real estate agencies actually use?

The practical stack for most agencies in 2026 is a general-purpose AI assistant on a business tier for drafting, one property data platform (Cotality RP Data, PropTrack or Pricefinder) for appraisals and CMA reports, and whatever AI features are built into their CRM. Rex, PropertyMe and Kolmeo all now ship AI features. Purpose-built listing tools like BoxBrownie AI Copywriting and AiRealEstate.Tech are also gaining adoption.

Do I need consent to use AI with buyer enquiry data?

It depends on how the AI processes the data. The OAIC's guidance confirms that entering personal information into a publicly available AI tool constitutes disclosure under APP 6. You need either explicit consent or a reasonable expectation from the individual that their data would be used that way. The safest approach is a clear AI disclosure clause in your agency's terms and privacy policy, particularly before 10 December 2026 when the automated decision-making transparency requirements commence.

Can AI-generated images be used in property listings?

Virtual staging is widely accepted with clear disclosure. Using AI to remove defects, add non-existent features, or misrepresent a property's condition is misleading conduct under the Australian Consumer Law. Queensland's Office of Fair Trading began targeted compliance checks on AI-altered listing images in August 2026. NSW Fair Trading already prohibits digital modification that changes a property's appearance. Disclose any enhancement, and never alter an image to conceal a defect.

Does my real estate agency need an AI policy?

Yes. From 10 December 2026, the Privacy Act will require agencies to disclose automated decision-making in their privacy policies. Beyond the legal requirement, an AI policy is also the governance step the federal Voluntary AI Safety Standard asks organisations to take.²⁶ It defines which tools are approved, what data can and cannot be entered, and who reviews AI-generated content before it goes out under the agency's name. The free AI policy template for Australian businesses covers all of these and aligns with the Voluntary AI Safety Standard's 10 guardrails.