👇 Should independent adjusters be using AI on their claims? Steve uses it on nearly every file he works, and in this interview he explains exactly how. He walks through dictating his inspections on site, training AI to write reports the way each firm wants them, using it to prioritize a full claims load, and the mistakes that will flag your reports as AI-written. Matt and Steve also talk about why every adjuster using AI has to become their own file reviewer, and why your license and reputation are on the line no matter what software wrote the words. Find adjuster training and Xactimate training at https://www.adjustertv.com/en
🎯 Want to Launch Your Adjusting Career? Stop begging for work. Most new adjusters wait by the phone while storms pass them by. 👉 Watch the free training now → https://adjustertv.com/deployment
If you want a clear, step-by-step plan to go from *zero to deployed* as fast as possible, start here:
HAAG EDUCATION: Use code ADJUSTERTV for a discount on books, trainings, certifications, continuing ed credits, and tools at: https://haageducation.com/
E&O insurance: Download the FREE GUIDE all about the different kinds of insurance you - as the adjuster - should carry: https://www.cplic.net/adjustertv/ ========================================================== 🧠 What is AdjusterTV? AdjusterTV helps you go from confused and stuck to confident, trained, and deployed as an independent insurance adjuster. Real advice from someone who’s worked in the field for 20+ years and helped thousands get started.
AI Generated·May contain errors
Video Summary
AI Generated
This video features a discussion between Matt and an experienced independent adjuster regarding the practical integration of AI in insurance adjusting. The guest explains how he uses AI as a productivity tool for report writing and claim management while emphasizing the critical need for human oversight to maintain professional reputation and legal accountability.
You've probably heard that adjusters
are starting to use AI to write their reports,
and maybe you're wondering if you should be doing it as well.
Or maybe you're worried that if you do,
a file reviewer is going to catch that,
kick your files back, or stop sending you claims altogether.
And if you're brand new, you might be wondering
if AI is going to make this job disappear
before you even get started.
I'm Matt, and I teach adjuster training
that gets experience requirements waived.
And in this video, you'll hear from a working independent adjuster.
who uses AI on just about every single claim he handles,
and he'll show you how he does it, where it goes wrong,
and how he keeps it from costing him his reputation.
And this video is brought to you by AdjusterTV.com.
If you want Adjuster training
or Xactimate certification training
from somebody who's actually written the estimates,
made a six-figure living from it,
that's where you're gonna find it, at AdjusterTV.com.
All right, let's jump into this, starting now.
Are you using AI? And if so, how are you using it with your clients?
Yeah, I mean, I'm using it a lot,
but as a tool, not as a output, you know, like.
Send my claim, boom, you know, and don't, you know, I,
the major way I use it, and I don't want to give all my trade secrets,
but, you know, one of my ways I use it, I think this is helpful
for the policyholder, so I should give it up, you know.
I dictate, so I just open up Apple Record, and I just spew,
and the policyholder's spewing, and we're just, we're just,
all this data is just getting collected by the voice recorder,
and at the end of my inspection.
It's not this easy. I had to train it to do this,
but I copy, transcript, paste, and boom,
whole report, front elevation, right elevation,
back elevation, cause of loss, underwriting
concerns. I spit out the number of stories,
just all the data you would need. And then from there,
it's not quite done yet, but I have all the Xactimate line items.
Built into it.
And I'm going to train it how to estimate the way I do.
And if I don't know, I'm going to say, hey, what do you think?
You know, and review it, you know,
instead of just searching exactimate blindly for what looks right.
You know, have you seen exact AI?
I've heard of it.
I've heard people using it.
I've seen some examples.
You know, I see drywall damage to a bathroom ceiling.
And the photo caption is a bathroom with toothbrush
and soap hole there on sink.
Well, so there's photo labeling, and then what you
were just describing is kind of what some of the newest,
I think it might still be in beta, but some of the newest features
that Veris is putting into Xactimate with AI are, they do just that.
They're like, you know, here's some other things you need
to consider for writing this estimate.
Or when you do a search, you can do a plain language search instead
of trying to like dig through their search tools.
And then it sends up, sets up, you know.
That line item suggestions and I think even totality is coming out
or has come out with you just say, replace the deck
and it just it spits it puts the line items into the estimate for you.
So yeah, I don't know. I haven't used those tools to be honest.
I, I think AI is kind of like drones.
They carriers don't necessarily they're not on board yet.
It's a liability more than it's a
They'd rather just have us work.
Yeah, or they don't have control over it.
Yeah, and then the T and E go, you know, especially hourly adjusters,
you know, it's how do you use it and justify the billing,
you know, it's a double-edged sword.
I think it's best at this stage
of the game just kind of being used as a calculator
in your pocket versus really openly using it as far as, you know.
Writing your full reports,
even though I use it to write reports, it's just every report's
different. Basically, if you train it right, you can use it.
But you have to train it like it's a
person. You have to say, this is the way this firm likes this report.
I always ask the firm,
can you send me your ideal report?
They always have one. The adjuster, eight years ago,
wrote this Bible on this claim.
And I put that in and I say, write it like this.
And and this is what I want. Don't say this. Don't say that.
It takes a lot of hours to train it. And then, boom, it does.
It spits it out almost perfect every time.
But you've got to use the right AI.
You've got to pay for it. You have to you know,
this you can't just go use the free plan and expect to get a good GLR.
You're going to get butchered.
You know, it's there's a lot of slop that you have to.
And then there's a lot of tendencies, like AI likes these long dashes.
When you see long dashes, you know it's AI.
You can't be putting that in your reports.
It's going to get the.
They're going to get sued, and they're going to lose.
Right.
I don't even know how to make an em dash with my keyboard.
I don't either.
I don't know.
So I don't know why it uses them.
Well, if you look, I'm actually looking at my, yeah, I don't know.
I don't know.
That's a good question.
That's a good point.
So really, so you're really using AI to help you write reports.
You've trained it to basically take your dictation of the report
and just clean it up.
So it's not really like it's doing the work.
I mean, it's basically at this point
with the training, it's doing the work.
You just got to make sure you check it.
Yes.
And make sure it's right.
You have to know the loss to make sure the report.
AI is great until you have to go in front of a judge.
Or a litigator, you know, you got to pretend like it's your license.
Most states are licensed.
The AI is not licensed. And if I had Joe write my,
Matt, hey, Matt, write my report for me. Here's
what happened. They're not going to take that.
They're not going to like that. So what's the
difference between you and a computer? So I think it's.
But if I ask you for advice, Matt,
how do you think I should write this? And you tell me,
I don't think that's bad as long as you can stand behind it and expect,
hey, this could be your
career on the line if you're wrong,
or the difference between the insurance company paying
for a floor, you paying for a floor.
And I use it other ways too. I use it to manage my whole,
I can go in and say, let's get to work.
And it says, it goes into my inbox. It goes into my
Claims board.
Cross-references emails with claims that this email is urgent
because this claim is urgent.
You should.
It's, it's pretty crazy. Yeah.
I can, cause you know, when you get 50 claims,
how do you triage that? Right. Right.
A lot of adjusters sit down and they're just like,
I think the game's on. I'll do this later.
The AI can prioritize it for you and say, you know,
Hey, uh, you should work on these claims,
but just so you know, these three claims haven't
been scheduled yet. And they've been sitting three days.
Uh, you should probably do that first.
Okay. You know, and then I do it. The way I train people is that they,
Before you turn your file, every single file,
I'm assuming you're not using AI.
Right. You have to go through your file and file review.
It takes two minutes, right? Just to
kind of like pull up the PDFs and just like scan it
and make sure there aren't, aren't like blanks
or the wrong F9 notes and pictures are out of order
or whatever it is. Right. Or you've got,
you know, you said there's depreciation in the F9 note,
but there's no depreciation on the
line items, et cetera. Things like dumb things
that file could get kicked back for that. They shouldn't
to waste time, save two minutes or take two minutes to save yourself.
It could be hours because if you make a mistake
that results in you having to change the estimate,
the dollar figure, you have to call the insured back, right?
And so now you've got phone tag.
So when you're using AI to do this kind of stuff,
you basically, you have to check it.
So basically, you become the file reviewer of the AI,
right, before you send it off to the other file reviewer.
So I think that it's incredibly important for people
to remember that no matter if you're using AI or not,
you got to double check your work before you send it in.
And I think it's, like you said,
it's probably even more critical that you check.
Read everything that it wrote, not just like you do a quick scan.
Every little word, because, you know,
I did one report where it said that it buried in
the report the cause of loss was determined
to be an unknown cause of loss.
Right.
You know, it's like that this doesn't,
you know, it's your reputation, especially when
you get the same file reviewer on all your files.
You know, when it's a different file reviewer every time,
you can kind of fly under the
radar for so long.
You know, when you're working for bigger, better firms
that have less people and, you know, you start to, you look like
an idiot if you say, if you don't check your work, you know,
so to your point, you know, you have to, you're the file reviewer now.
I mean, you're writing the reports too.
You're telling it what to write.
I mean, it's not just like, although I can upload photos
and have it analyze what's in the photos and I can do that.
That's not how it gets the info.
It gets the info from me on site when I'm dictating that dictation.
When I'm talking to it, I'm talking to it like it's actively,
I get someone at home writing a report.
And I say it out, and then it makes it sound nice.
Not like a crazy New Englander who talks like crazy.
Obviously, you started in 22.
AI was a thing back then, but I don't think it was like,
you probably weren't using it for claims back then.
So how would you say that AI has helped you?
Because you're using it, it's obviously,
you know, you wouldn't be using it if it wasn't
helpful, if it didn't make you faster.
Like, what would you say? How would you say it's been
helping you? I use the analogy that, you know.
If, if explosions were technological advancement,
you know, the telephone was a firecracker.
The internet was a, you know, was a,
a container of a giant, one of those liquid natural gas
container, you know, it was pretty big explosion.
AI is like a nuke.
I mean, it's insane.
It's, it's almost like the internet was always meant to be AI.
It just wasn't developed yet.
The beast isn't done being created from the 80s and 90s
when we were just going in chat rooms
and sharing pictures of guitars and stuff.
You know, you know.
And now it's just that data that's all been massed on the internet.
So I have a lot of e-books.
And now I can just dump those e-books in a folder
Ask questions and it'll read all those books
and give me the context.
It's not Google giving me the info.
If you have an honest AI, you'll have honest answers.
That's so funny that you said that.
My sort of thing that I said when AI looked like
it was going to be kind of a big deal
and definitely a nuclear for sure
is that the computers are finally starting to do
what the promise of them was originally, right?
You just tell it what you want in plain language
And it's, can.
Now with agents and all that stuff,
I mean, it goes out and buys stuff for you and whatever.
And who knows?
It's like two years from now.
I mean, good grief.
If you're wondering whether drones and AI
are coming for the adjuster job altogether,
Steve's got some strong opinions on that
in this video right here.
And as always, thank you so much for watching
and have a great storm.