Updated 19 September 2026: first published on Elite Agent on 9 April 2026. I’ve revised the video-assessment, AI shortlisting and privacy guidance to focus on job-related evidence, fair assessment and meaningful human review.
Everyone’s CV looks amazing now. Here’s how to find the actual human behind the prompt.
Thomas McGlynn nailed it a few years back when he said that a job interview has always been the meeting of two lies: The employer exaggerates how great the culture is. The candidate exaggerates how great they are. Ninety days later, everyone finds out.
Now multiply that problem by AI.
Today’s candidates are running their cover letters through ChatGPT. Their CVs have been polished by Gemini. Their LinkedIn summaries read like they were written by a professional copywriter – because, in a way, they were. And here’s the thing: you can’t really blame them. If you had a tool that could make you sound more articulate, you’d use it too.
But it means the old hiring playbook could be a little broken.
A polished cover letter can tell you what someone wants you to know, but it needs backing up. And dumping AI-polished applications into another AI without checking its summaries? Congratulations – you’ve just created a game of robot tennis where nobody’s keeping score and the human wandered off to make a coffee.
So what do you do instead?
Change the medium, change the signal
When polished writing is easy to produce, a written application becomes a starting point. You still need to check the experience and skills behind it.
Give candidates a chance to show how they work, then ask them to explain their decisions.
Build a smarter application form
Build situational questions into your application form, using realistic examples from the role:
- A tenant calls at 4:45 PM on a Friday with a burst pipe. Walk us through exactly what you do in the next 30 minutes.
- A seller disagrees with your recommended sale price. How do you handle the conversation?
- You’ve got three inspections, a settlement, and a staff meeting all on the same morning. What gets moved?
AI can help answer these scenarios too. The useful part comes in the follow-up: why that action, what would they check, and what would change their mind? Give everyone the same core questions and assess the answers against the same job-related criteria.
Ask about adaptability and task preferences
Ask candidates about a time they had to learn a new system or change their approach. What did they try, and what happened? You can also ask them to rate how much they enjoy different tasks. If someone rates admin a 2 and the role is mostly systems and data entry, that’s a conversation to have. A preference score is a discussion starter, not a validated measure of adaptability or a reason to reject someone on its own.
Think of it as a conversation about the actual working day. Tinder for task preferences, minus the awkward ghosting. (Actually, no – realistically there’s probably still ghosting!)
Use video as a conversation starter
A short video answering a role-related question can add context, particularly when communication is part of the job. But it isn’t proof of authenticity or character: AI can generate convincing faces, voices and video. And confidence on camera isn’t the same thing as competence at work.
Keep the question simple: Tell us why this role, at this agency, right now. Then follow up in a live conversation or a short, relevant work sample. Fifteen seconds is a first impression, not a hiring decision. Offer an equivalent phone or live-interview option and reasonable adjustments. Assess the substance of the answer, not someone’s appearance, accent or facial expressions.
Ask the questions that actually matter
Once you have some evidence of their experience and approach, use the interview to explore it.
The culture question nobody asks:
Most agencies advertise the same role. Same duties, same salary band, same “dynamic team environment.”
Your culture matters – so explain the working practices and expectations behind that word.
Give candidates the same links to your agency’s public work and ask what they noticed. Then ask how they would handle a realistic team situation. You’re looking for evidence relevant to the work, not whether someone shares your interests or looks like the existing team.
The future question:
If property management shifts from being mostly about maintenance to being mostly about asset advisory and client experience, how would they see their role changing?
Use the answer to explore what they would need to learn and how they would approach the change.
The AI question:
Ask: Where would you use AI in this role, and where would you check its work or leave the decision to a person? Follow up with an example of something they have learned. Curiosity matters, and so does knowing when to be cautious.
Keep assessments tied to the work.
If you use a profiling or psychometric tool, check the evidence that it is suitable for the role and hiring purpose, and get qualified help interpreting it. A personality label shouldn’t decide the shortlist. A structured interview, a relevant work sample and references give you concrete evidence to discuss.
Use AI on your side (smartly)
Here’s where it gets fun. AI isn’t just changing how people apply – it can transform how you hire.
Use AI to help you build your application form in the first place. Give it a role description without personal or confidential details and ask for situational questions. Review them yourself for relevance and fairness before using them.
It can also help draft a scoring rubric before applications arrive. You decide the job-related criteria and what good evidence looks like. If an approved tool summarises applications, have a person check each summary against the original and review the shortlist before invitations or rejections go out. Send the booking link after that review.
The workflow becomes: application lands → AI helps organise the responses → you check the evidence and decide who to interview → you walk into the conversation prepared. You know, the bit you as a leader went into real estate for: talking to people. Not reading 47 variations of “I’m a passionate self-starter.”
That only helps if the review is real. Keep the original responses available, check for omissions and inconsistent treatment, and record the reasons for your decisions. Don’t ask AI to infer honesty, personality or suitability from a candidate’s face or voice.
The bit about staying legal
As at 17 September 2026, new automated-decision transparency requirements are due to start on 10 December 2026. The OAIC’s guidance sets out their scope: they apply to organisations and agencies covered by the Australian Privacy Principles when a computer program uses personal information to make, or do something substantially and directly related to making, a decision that could reasonably be expected to significantly affect someone’s rights or interests.
AI-assisted shortlisting may fall within that scope. Having a person make the final hiring decision does not automatically take the process outside it. Where the requirements apply, the privacy policy must describe the kinds of personal information used and the relevant kinds of automated or computer-assisted decisions. Check how this applies to your agency and its actual process.
Three practical places to start:
Explain the process. Tell applicants what the AI actually does and where a person reviews the work. If it only summarises responses, say that. If it scores or ranks them, explain that too. Give applicants a contact for questions or corrections.
Choose the tool before using applicant data. Check access, retention, overseas handling and whether information is used for model training. The OAIC recommends keeping personal information out of publicly available generative AI tools. Use fictional examples while designing the process.
Make the policy match the practice. Review your privacy notice, policy and data handling with the people responsible for recruitment and privacy. A generic “we use AI” sentence doesn’t explain the whole process.
This isn’t something to panic about. It’s something to get right before you rely on it. Transparency matters, and so do fair criteria and meaningful human review.
These are starting points, not a complete compliance check. Existing privacy and anti-discrimination obligations still matter. The Australian Human Rights Commission’s AI and recruitment checklist is a useful resource to work through with your adviser.
Go beyond the resume. Look for evidence.
The job interview was always the meeting of two lies. AI has just given both sides a better script. Your job as a hiring manager hasn’t changed – find the right person for your team – but your methods need to catch up.
Ask people to explain how they work. Test the skills the role actually needs. Give them a fair chance to demonstrate those skills, and use AI to help you prepare and organise.
That gives you a better basis for the conversation. And maybe, just maybe, both sides will be telling the truth by interview two.
Ready to find your next team member? Post your role on Elite Agent Jobs – it’s free to post, with optional promotion if you want more reach.
For more on applying AI to the systems in your business, explore the AI First Agent Course, which teaches the AI First Formula and practical real estate workflows.
Read the refreshed version on AI Powered Real Estate by Samantha McLean on Substack.