Skip to content
RICS AI Standard Aligned

AI that a regulated professional can defend.

PF & Co Site Intelligence is not itself RICS-regulated, but our practices are mapped to the RICS Responsible use of AI in surveying practice standard (mandatory from 9 March 2026). This page is the public evidence your RICS-regulated clients will need for supplier due diligence.

The RICS standard, mapped to our practice

The RICS standard sets out four key provisions. Below is how each maps to our documented practice. Your RICS-regulated clients can cite this page when recording their supplier due diligence.

01

Governance & Risk Management

RICS requirement

Firms must set clear policies on data use and AI system governance, keep a written register of material-impact AI systems, and maintain a risk register (documenting bias, erroneous-output and data-quality risks), reviewed regularly.

Our practice

We maintain an internal AI register and a risk register covering bias, erroneous-output and data-quality limits, reviewed regularly. Client data is processed solely to produce the ordered report and is never used to train or improve AI models. UK GDPR / DPA 2018 compliant handling. Published Privacy Policy and AI Governance Policy.

02

Professional Judgment & Oversight

RICS requirement

A named, appropriately qualified surveyor must assess the reliability of AI outputs and remain accountable for the work, applying professional scepticism throughout; automated or high-volume use is dip-sampled.

Our practice

Every report passes 200+ automated quality checks across accuracy, regulatory compliance, consistency, and citation integrity before issue, and Site Intelligence stands behind every brief it issues. For clients who want documented professional scrutiny, an optional add-on provides a named personal sign-off or a chartered-planner review of the full pack; because the reviewer reads the whole brief, it adds time beyond the 48 hours.

03

Transparency & Client Communication

RICS requirement

Clients must be told in writing, in advance, when and how AI is used, the extent of any professional-indemnity cover for that use, and how to contest it or seek redress; firms must be able to explain the system on request.

Our practice

This public AI Standards & Compliance page, AI disclosure at the point of order, and an AI methodology section in every report. Client information and redress rights are set out in our Terms of Service (Section 7.3), and we can provide an explainability summary on request.

04

Ethical Development of AI

RICS requirement

Firms that develop their own AI must assess training-data quality, relevance and diversity (and known gaps and bias), involve diverse stakeholders, consider sustainability impact, and comply with data and confidentiality law.

Our practice

We build our own AI system on authoritative government, statutory and regulated-industry data only — no unverified web scraping, social media or crowd-sourced data as primary inputs. We assess data quality and known gaps, manage bias in constraint screening and scoring, and process all data lawfully.

Operational safeguards

The specific controls that translate the standard into daily practice. Automated quality checks are standard on every brief; a named or chartered human sign-off is an optional add-on.

24-category QA pipeline

Every report passes 200+ automated checks across accuracy, regulatory compliance, consistency, formatting, and citation integrity before issue. This is standard on every order, within the 48 hours.

Optional expert sign-off

For clients who want extra assurance, a named personal sign-off or a chartered-planner review of the full pack is available as an optional add-on. The reviewer reads the whole brief, which is why it adds time beyond the 48-hour turnaround.

No unchecked delivery

There is no pathway by which a report reaches a client without clearing the full automated quality pipeline. Every brief is checked before issue, every time.

Authoritative sources only

130 authoritative government, statutory and regulated-industry datasets. No unverified web scraping or social media data.

Client data isolation

Your site data is used only for producing your ordered report. It is never used to train or improve AI models.

Traceable outputs

Every number in every report traces back to its authoritative source. If a planning officer asks "where did you get this?", one click reveals the answer.

Supplier Due Diligence Pack

A detailed document for RICS-regulated professionals fulfilling their supplier due diligence obligations. Covers AI systems summary, data source categories, analysis domains, QA pipeline, oversight arrangements, data governance, and the full RICS standard alignment mapping.

Read the pack

Current version: 1.0 (13 March 2026). For a PDF copy or questions about our AI governance: info@pfandco.co.uk.