Aipedia: Quantity surveyors

The RICS AI standard applied to quantity surveying work

Aipedia, by the Apex broking team · Last reviewed 2026-10-30

The RICS AI standard applies to quantity surveyors in RICS-regulated firms exactly as it does to other surveyors, and has been mandatory since 9 March 2026. Cost tools feed advice that clients use to fund, pay and settle, so many QS uses of AI may need its full set of records and checks.

Part of: AI in quantity surveying

In short

The RICS professional standard on responsible use of AI in surveying practice, published September 2025, is binding on RICS members and regulated firms from 9 March 2026. For quantity surveyors, the key duties are a written material-impact record for estimating and take-off tools (1.2), written due diligence on cost-data and estimating vendors (4.1), and a written reliability decision by a named, appropriately qualified surveyor, with dip sampling of automated rates and quantities (4.2). Clients must be told in advance (4.3). RICS has issued no QS-specific AI guidance, so firms apply the general standard to cost work themselves.

What RICS has published and its status

RICS is the professional body for chartered quantity surveyors and RICS-regulated firms. There is no statutory regulator of quantity surveying, so RICS standards are the main source of binding professional rules on AI.

Document	Date	Status	Relevance to QS work
Responsible use of AI in surveying practice (1st edition)	Published September 2025; in effect 9 March 2026	Mandatory professional standard	Applies to all RICS members and regulated firms worldwide, including QSs
Rules of Conduct	Effective 2 February 2022	Binding	Rule 2 competence and Rule 3 diligent service underpin the AI duties
AI in Construction 2025 report	16 September 2025	Research, not guidance	Survey of adoption, barriers and attitudes in construction
Modus article on AI and expert witnesses	24 November 2025	Commentary	Relevant to quantum experts

RICS has not published guidance specific to AI in quantity surveying, for example on AI estimating or AI-assisted measurement. QS firms therefore have to work out how the general standard applies to their own tools. This page does that for cost work; the firm-wide steps are in our RICS AI standard compliance checklist.

Section 1.2: the material-impact record for cost tools

Much of the standard applies where AI output has a "material impact" on the service. For a QS, that will usually include automated take-off, AI estimating and benchmarking, valuation tools and generative AI used to draft claims, because their outputs feed advice a client relies on. The firm must record any material-impact determination in writing, with reasons.

In practice, keep a short note for each tool: what it produces, which deliverables it feeds (cost plans, valuations, final accounts, claims) and why the firm decided the impact is or is not material. A tool used only to tidy the formatting of a report may not be material; a rate library that populates a pre-tender estimate almost certainly is.

Section 4.1: due diligence on cost-data and estimating vendors

Section 4.1 requires written due diligence questions to AI suppliers, with the answers recorded. For a QS, the suppliers include cost-data platforms, estimating software and take-off tools as well as general drafting assistants. Useful questions include:

Where the cost data comes from, how old it is and how often it is updated.
How location, inflation and procurement adjustments are applied, and whether they can be seen and overridden.
Whether the supplier trains its models on client data uploaded by users.
What the supplier accepts liability for if the output is wrong.

If a vendor cannot explain how a benchmark is built, record that and treat its output with more scepticism.

Section 4.2: named sign-off and dip sampling of automated rates

Section 4.2 requires professional scepticism and a written decision on whether an output is reliable, made by or under the supervision of a named, appropriately qualified surveyor. For high-volume or automated outputs, the standard expects dip sampling.

For cost work this means:

A named QS signs off each cost plan, estimate, valuation or final account built with AI, with a short record of what was checked.
Automated take-off is dip sampled against the drawings, with particular attention to scope that is not drawn.
AI-suggested rates are sampled against recent tender returns or the firm's own knowledge of the market.
Where sampling finds errors, the sample is widened and the issue logged on the AI risk register (3.3).

The other duties in a QS setting

The standard's other duties apply to QS work as follows:

Section 2, knowledge: members need a baseline understanding of AI limits, erroneous output, bias and data risks. For a QS, that includes knowing how historic cost data can mislead.
Section 3.1, confidentiality: no uploading of tender prices, client budgets, contractor rates or claims material to an AI system without express prior written consent, and at least annual staff training. See AI tools and tender confidentiality.
Sections 3.2 and 3.3, registers: a written assessment that AI is the right tool, an AI register (system, purpose, first use, next review) and a risk register reviewed at least quarterly.
Sections 4.3 and 4.4, clients: tell clients in writing and in advance when and why AI is used, with contest, redress and opt-out processes and, if available, the extent of PI cover for AI use; explain the system on request.

What the construction data says

The RICS AI in Construction 2025 report drew on more than 2,200 professionals. It found 45% of organisations reported no AI use and only 1% had scaled AI across projects. Barriers were a lack of skilled personnel (46%), system integration (37%) and poor data quality (30%). Nearly 70% of project managers and quantity surveyors believe AI will help them deliver greater value, and 25% plan to increase AI spending in the next 12 months.

The data quality finding matters most for compliance. If the underlying cost data is weak, the 4.1 and 4.2 records are where a firm shows it noticed.

What it means for PI

The RICS Approved Minimum Wording does not mention AI, and the standard does not change the RICS PII minimums. But CMS has said the standard is likely to inform how negligence claims are judged. A firm that can show a material-impact record, vendor questions and a named sign-off is in a stronger position to show reasonable skill and care. Gaps in those records are likely to be used against it. Read more in the RICS AI standard and PI and whether QS PI covers AI cost errors. Apex arranges PI for quantity surveyors and can check how your cover and proposal answers reflect the standard.

Sources

Frequently asked

Does the RICS AI standard apply to contractor-side QSs?

It applies to RICS members and RICS-regulated firms, so a chartered QS working for a contractor is bound by its member duties. Firm-level duties such as registers fall on RICS-regulated firms. Non-RICS contractor QSs are not bound, though their employer may adopt similar controls.

Is estimating software that has been used for years now caught?

Only if it falls within the standard's scope as an AI system and its output has a material impact. Traditional rule-based software may fall outside, but the firm should still record its reasoning. Tools that now add machine-learning rate suggestions or automated recognition are more likely to be caught.

How much dip sampling is enough?

The standard does not set a percentage. The sample should be enough for a named, appropriately qualified surveyor to make a defensible reliability decision, larger where the output is high value or the tool is new, and widened if errors are found.

Has RICS issued QS-specific AI guidance?

RICS had not published QS-specific AI guidance as at October 2026. Quantity surveyors apply the general standard, published September 2025 and mandatory from 9 March 2026, to their own estimating, valuation and claims tools.

Check your PI covers how you use AI

Tell us about your firm and the work you do. A broker will review how your current PI treats AI-assisted work, or arrange new cover. No fee for a quote. Or call 0117 325 0027.

Start a PI quote Request a callback

Apex Insurance Brokers Limited is authorised and regulated by the Financial Conduct Authority. Registered in England and Wales, company number 07014570. This page is general information about how AI is used in the profession and how insurance may respond. It is not legal, regulatory or insurance advice on your circumstances, and it does not guarantee that cover will be available or on what terms.