DON'T DO THIS! AI Is a Tool, Not a Shortcut: What I Learned at an AI Risk Management Workshop
Last week I sat in on an AI risk management workshop, and I walked out thinking about it differently than I walked in.
I use AI every day. It helps me draft market reports, organize my notes, and build tools that make my clients' lives easier. I'm not here to tell you to stop using it. I'm here to share what I learned about using it well, because the risks aren't just a real estate problem. If you're a lawyer, an accountant, a nurse, a contractor, or a small business owner, the same rules apply to you.
Here's the big question the workshop kept circling back to: what are you feeding your AI, and what are you doing with what it gives back?
First, Let's Talk About AI Fatigue
You've felt it. You scroll past a post and something about it just feels… off. The phrasing is too polished, the image is a little too perfect, and you can tell it took someone about eight seconds to make. People call it "AI slop." I call what it's causing AI fatigue, and it's real.
The numbers back it up. A Gartner survey from this summer found that 65% of consumers think brands are putting out too much AI-generated content, and 57% say it's made them trust brand messaging less. Sprout Social found that half of Gen Z has already blocked, muted, or unfollowed a brand or creator because their content felt like AI slop.
Here's what I think is really going on. People are making a quick judgment before they decide to give you their attention. If it looks like you didn't put much into it, why would they put much into reading it? And the leap doesn't stop there. If your content looks rushed, people start wondering what else you're rushing.
I'll be honest with you. My website platform started auto-generating blog drafts for me a while back. Some of them weren't bad! But none of them sounded like me, and none of them knew anything about Avon, or what's actually happening on the South Shore. So I don't post them as-is. I take the idea and rebuild it with real local detail and my own voice. That's the difference between using AI as a tool and letting it do your talking for you.
AI isn't the problem. Effortless content is the problem. The people who win will be the ones who use AI to do more thoughtful work, not less work.
Risk #1: Be Careful What You Feed It
This was the biggest wake-up call from the workshop. Depending on the tool and your settings, what you type into an AI can be stored, reviewed, and even used to train future versions of the model. The safest mindset? Treat a public AI chatbot like a public space. If you wouldn't say it out loud in a crowded coffee shop, think twice before pasting it in.
For me, that means a client's financial details, a seller's reason for moving, a buyer's offer strategy, or anything from a transaction file doesn't get dropped into a random AI tool. My clients trust me with that information. Protecting it is part of the job.
A few habits that came out of the workshop and my own digging:
Check your settings today. Most major AI platforms have a privacy or data-controls setting that lets you turn off the option to use your chats for training. It takes two minutes.
Use the approved tools. If your company, brokerage, or firm has vetted specific AI tools, stick to those for anything work-related. Business and enterprise versions typically come with stronger privacy guarantees than free versions.
Strip out the names. Need help with a tricky situation? Describe it without names, addresses, account numbers, or anything that identifies a real person.
Watch the AI note-takers. Those apps that join your video calls and summarize them are handy, but everything said in that meeting is now sitting on someone's server. Know what you're agreeing to, and let others on the call know it's there.
Never, ever paste in Social Security numbers, bank details, passwords, or medical information. Full stop.
Risk #2: Don't Trust It Blindly
AI is confident. That's the problem. It will hand you a wrong answer in the same polished tone as a right one. The industry calls these "hallucinations," and they've already caught people off guard. Lawyers have been sanctioned in court for filing briefs with case citations an AI made up. The AI sounded sure. The cases didn't exist.
In my world, the risks look like this:
Accuracy. A wrong square footage, a made-up school stat, or an outdated tax rate in a listing isn't just embarrassing. It can cost a client money and cost you your reputation. Verify every fact with a real source.
Fair housing. AI learned from decades of internet writing, and not all of it was fair. It can slip in language that describes who should live somewhere instead of what the home offers. NAR's guidance is clear that AI content still has to meet your duty to be truthful and present a true picture, and you're responsible for it as if you wrote every word yourself.
Images that mislead. AI can "improve" a photo right into a feature the house doesn't have. A virtually staged room should be clearly labeled as staged. If a buyer walks in and feels tricked, you've lost them before they reach the kitchen.
Originality. AI can produce content that sounds a lot like someone else's work. Rewrite it in your own words, with your own experience. That's not just about avoiding trouble. It's how your content stops sounding like everyone else's.
The simple rule I took home: AI drafts. You decide. Your judgment is the value. Clients can get generic words anywhere. They come to you for the part a machine can't do.
Risk #3: The Scammers Are Using AI Too
This is the one that hits closest to home for my clients. The FBI's 2025 Internet Crime Report found that real estate fraud reached 12,368 complaints and $275 million in losses, a 59% jump from the year before. For the first time, the FBI also tracked AI specifically, logging more than 22,000 complaints tied to AI and $893 million in losses.
And here's the part that gave me chills. Scammers can now clone a voice from a short clip. Earlier this fall, CNN reported on a couple who lost $66,000 in closing funds after hearing what sounded like their own mortgage professional. It wasn't him.
This isn't only a real estate problem, either. Voice cloning has made the old "grandparent scam" far more convincing, and deepfake videos are being used to fake endorsements from trusted figures.
What you can do:
Verify through a second channel. Got an email, text, or call asking you to send money or change payment details? Hang up and call back using a number you already know is real, not the one in the message.
Wire instructions don't change by email. If you're buying a home and suddenly get "updated" wiring instructions, stop. Call your title company or attorney directly. Every time.
Recognizing the voice isn't enough anymore. Even if it sounds exactly like someone you know, double-check before acting on anything financial.
Set a family code word. A simple word only your family knows can stop a fake emergency call in its tracks.
Lock down your accounts. Strong, unique passwords and two-factor authentication on email are your first line of defense. Most of these scams start with a hacked inbox.
Move fast if something goes wrong. Call your bank immediately and file a report at ic3.gov. With wire fraud, every minute counts.
A Quick Gut-Check Before You Hit Enter (or Publish)
Whether you sell homes, manage a practice, or run a small business, these five questions cover most of it:
Would I be comfortable if this prompt showed up somewhere public? If not, take out the private details.
Am I using a tool my company approves, with training turned off?
Did I verify every fact, number, and claim with a real source?
Does this sound like me, and does it reflect real effort? If it reads like it took eight seconds, your audience will know.
If money is moving, did I confirm it through a second channel?
Let's Keep This Conversation Going
AI is going to keep getting better, and I'm going to keep using it. But the workshop reminded me that the tool is only as trustworthy as the person using it. Your clients, your customers, and your audience aren't trusting an AI. They're trusting you.
So I'm curious. Has AI fatigue changed how you scroll, read, or decide who to trust? And if you've ever gotten a call or email that almost fooled you, I'd love to hear how you caught it. Share it in the comments. Your story might be the one that helps someone else spot the next scam.
And if you're buying or selling and want to talk through how to protect yourself during a transaction, I'm always happy to grab a coffee and walk you through it. No pressure, just a conversation.
Comments