Your report is already inside ChatGPT.

Surveyors UK

Blog

Technology & AI

27/02/2026

ChatGPT and survey reports: the short version

Clients and rival firms are putting survey reports into ChatGPT and using what comes back to challenge the surveyor. A report that took two days to write can produce follow-up questions, challenges and a draft complaint letter in seconds. The cost to challenge has dropped close to zero. The cost to defend has not moved. One surveyor described a rival firm uploading their report into ChatGPT to critique it: the AI cited legislation that does not exist, the report was under an NDA, and confidential data went into a public tool. Hallucinated law, a data breach and a contract breach in a single incident. Terms of engagement written before ChatGPT existed do not address any of it.

Key takeaways

Clients and rival firms are already uploading survey reports into ChatGPT to generate challenges and draft complaint letters
The cost to challenge a report has fallen close to zero. The cost to defend one has not changed
In one reported incident, a rival firm’s ChatGPT critique cited legislation that does not exist
Putting a report that is under an NDA into a public AI tool is both a data breach and a contract breach. What this means for your PI cover
Terms of engagement written before ChatGPT do not say whether clients may upload your report into public AI tools. What the RICS AI standard requires

This week, I’m focusing on one topic because something happened this week that I think every surveyor and firm owner needs to read.

Before I get into it, a quick update. We’re building out dedicated spaces for surveying firms inside Surveyors UK this spring. Directory profiles, referral network access, job posts, and more. Over 36 firms are already on the waitlist.

Now to the thing that’s had my inbox buzzing.

Your client just uploaded your report into ChatGPT

I posted something on LinkedIn this week about AI and complaints. The response told me something important. Most of the profession hasn’t fully grasped what is unfolding here. Until it happens to your firm, it’s easy to assume this is a future problem. The comments and DMs proved otherwise. Surveyors are already dealing with this. They just hadn’t connected the dots until now.

Read the post and comments here

The scenario is simple. A surveyor spends two days on a report. Detailed, professional, and considered. The client uploads it into ChatGPT and, within seconds, has a list of follow-up questions, potential challenges, and a draft complaint letter.

The cost to challenge is dropping to almost zero. The cost to defend hasn’t changed at all.

But it was the comments that really opened this up. Surveyors sharing real experiences, raising questions nobody in the profession seems to be asking yet, and starting conversations that are long overdue.

A real incident. Three problems in one story.

One surveyor shared what happened when a rival firm uploaded their report into ChatGPT to critique it. The AI cited legislation that doesn’t exist. The report was subject to an NDA. And confidential data was placed into a public AI tool without a second thought.

Hallucinated law. A data breach. A contract breach. All in one incident.

This isn’t a theoretical risk. It is happening right now.

And it’s not just rivals. This week the Upper Tribunal ruled on a solicitor who admitted uploading client documents into ChatGPT. The judge was clear. It places confidential information into the public domain, breaches client confidentiality, and waives legal privilege. The solicitor is now facing an investigation.

That ruling was about a professional uploading client data. The profession hasn’t even started talking about what happens when clients upload professional reports. Reports containing personal information, property details, and sensitive findings. No consent. No controls. No oversight.

And it won’t always be the instructing client. What happens when someone gets hold of a report that isn’t theirs and drops it into an AI tool? This is happening NOW.

The cost to challenge versus the cost to defend.

Another comment cut straight to the commercial reality. It takes a client minutes to generate a complaint using AI. It costs the firm weeks of time, real money, and serious stress to investigate and respond. And that’s before you get into the legal and PI aspects.

That imbalance is only going to grow. AI is making it cheaper and easier to challenge professional work. But defending that work still costs exactly what it always did.

Every firm owner reading this will have felt that in their gut.

So what do firms actually do about it?

One of the most practical comments suggested adding AI-specific clauses to terms of engagement. An exclusion preventing clients from uploading reports into public AI tools, with clear consequences for breach.

It’s worth exploring. Any clause like this would need proper legal review, especially around enforceability under consumer contract rules. But the instinct is right. Terms of engagement were written for a world that no longer exists. They need updating.

A single clause isn’t enough on its own, though. Firms need a structured approach to how AI sits within their practice. How they govern it. How they document their reasoning. How do they evidence their professional judgement when it’s challenged?

This is why I built the GUARD Framework.

Governance. Use. Accountability. Risk. Documentation. Five pillars designed to help surveying firms start thinking about AI governance practically and proportionately.

It won’t solve everything. No single framework can. But it gives firms a structured starting point for the conversations they need to be having right now, along with the documentation to back them up.

On 31 March, I’m running a practical session to walk firms through the framework and give you tools you can start using immediately. Over 20 places already taken.

Save your spot here.

The profession is moving fast. Make sure your governance is moving with it.

Until next week

Nina

Frequently asked questions

Are clients really putting survey reports into ChatGPT?

Yes, and so are rival firms. A report that took two days to write can produce follow-up questions, challenges and a draft complaint letter in seconds. Surveyors are already dealing with this. Many had not connected the dots until it happened to them.

What actually goes wrong when a report is put into ChatGPT?

One surveyor described a rival firm uploading their report into ChatGPT to critique it. The AI cited legislation that does not exist. The report was subject to an NDA. Confidential data was placed into a public AI tool without a second thought. Hallucinated law, a data breach and a contract breach, all in one incident.

Why does this change the economics of a complaint?

Because the cost to challenge is dropping to almost zero, while the cost to defend has not changed at all. A client who would previously have needed to instruct someone to interrogate your report can now do it in seconds for nothing.

Can I stop clients uploading my report into AI tools?

Terms of engagement were written for a world that no longer exists and need updating. One option worth exploring is a clause preventing clients from uploading reports into public AI tools, with clear consequences for breach. Any clause like that would need proper legal review, particularly around enforceability under consumer contract rules.

Is a contract clause enough on its own?

No. A single clause is not enough. Firms need a structured approach to how AI sits within their practice: how they govern it, how they document their reasoning, and how they evidence their professional judgement when it is challenged.

Nina Young