top of page

Where AI Fits in a Property Media Workflow (and Where It Doesn't)

Writer: Jonathan G. Browning
Jonathan G. Browning
Oct 7
3 min read

AI is showing up in every part of property work: tools that sort photos, write listing descriptions, replace skies, stage empty rooms, and draft reports. Some of it saves real time. Some of it creates a problem that did not exist before: an image or a document that looks like a record but is not one.

The question is not whether to use these tools. It is which jobs to hand them, and how to keep a person responsible for what goes out the door.


Separate the busywork from the judgment

Most of the time in a property media job is not spent shooting. It is spent on everything around the shoot:

  • Collecting the address, access notes, and goals from the client.

  • Renaming, sorting, and backing up files.

  • Picking a first round of selects from hundreds of frames.

  • Formatting deliverables and sending notices when they are ready.

These are good candidates for automation. They are repetitive, the rules are clear, and a mistake is easy to catch.

Judgment calls are different. Deciding whether a crack in a foundation belongs in the record, whether an edit crosses a line, or what a client should be told about a property are human decisions. Automation can prepare the work. A person should make the call.


Use a framework, not a feeling

The National Institute of Standards and Technology publishes the AI Risk Management Framework. NIST describes it as intended for voluntary use, to help organizations incorporate trustworthiness considerations into the design, development, use, and evaluation of AI products, services, and systems. It was released on January 26, 2023, and NIST notes that it is being revised.

You do not need to adopt a federal framework to run a small business. But the idea behind it is useful: name the risks before you use a tool. For property work, three questions cover most of it.

  1. What happens if it is wrong? A misnamed file is an annoyance. A wrong acreage figure in a listing is a problem.

  2. Who checks the output? Every automated step needs a named person who reviews it.

  3. Does the result change what a viewer believes about the property? If yes, it needs a label.


Laptop and phone on a dark desk with code open on the screen

Records and renderings are different things

A photograph of a room records what was there when the shutter fired. A virtually staged version of that photo shows a possibility. Both are useful. They should never be confused.

The National Association of Realtors Code of Ethics, Article 12, says Realtors shall present a true picture in their advertising, marketing, and other representations. You can read it in the NAR Code of Ethics and Standards of Practice. AI editing makes it easier than ever to drift away from that true picture without meaning to.

The fix is labeling. Any image where AI added, removed, or replaced content should say so on the image or right next to it. "Virtually staged." "Sky replaced." "Concept rendering." Short and plain.


Provenance is catching up

There is also a technical side to this. The Coalition for Content Provenance and Authenticity, or C2PA, provides an open technical standard for establishing the origin and edits of digital content. C2PA calls it Content Credentials and compares it to a nutrition label for digital content: a history of the file that anyone can check.

Support for Content Credentials varies by camera, software, and platform. It is worth watching. For now, the practical steps are simpler: keep original files, keep edited files separate, and keep a short log of what was changed.


A sensible starting point

If you want to bring automation into a property workflow, start small:

  • Standardize your intake form so every job starts with the same information.

  • Automate file naming and backup.

  • Let software suggest selects, then review them yourself.

  • Draft routine messages with a template, and read every one before it goes out.

Each step removes a little friction without handing over a decision that should stay with a person.


Practical takeaways

  1. Automate repetitive tasks with clear rules. Keep judgment calls human.

  2. Assign a person to review every automated output.

  3. Label any image where AI changed the content.

  4. Keep originals separate from edited files.

  5. Use a simple risk check before adopting a new tool: what if it is wrong, and who would notice?

If you are working through how documentation and automation fit your property or project, contact 48 Pegasus, serving properties within 75 miles of Macon County, Tennessee, across Middle Tennessee and South Central Kentucky.


Sources

Photos: cover by Thanos Pal / Unsplash; inline by Bayu Syaits / Unsplash. Free to use under the Unsplash License.

Comments

bottom of page