12 May 2026
Controlled adoption, not blanket bans
Ask around the built environment and you will hear two extreme positions on AI. One camp wants it banned from anything professional. The other is pasting confidential tender information into free chatbots without a second thought.
The official UK guidance, from professional bodies, regulators and government, supports neither. Across the sector the prevailing approach is managed use: keep professional judgement in the loop, control your data, document decisions, and build governance and training so risks are caught early and checked routinely.
What the institutions actually say
RICS has published a professional standard on the responsible use of AI in surveying practice, effective March 2026. It requires risk registers, documented reliability decisions for material outputs, and proportionate assurance. Notably, it does not demand that firms check every output: for high-volume automated use it says that is generally neither necessary nor proportionate, and requires randomised dip-sampling instead, with the firm remaining accountable.
CIOB's AI Playbook positions the technology as capable of improving skills, sustainability, quality and safety, while flagging the risks to manage: data quality, IP and GDPR, and contractual clarity about expectations and risk allocation.
RIBA's 2025 AI report found practices using AI rose from 41 per cent in 2024 to 59 per cent in 2025, alongside recorded concerns about imitation and reliability. Adoption is happening either way; the question is whether it is governed.
The Institution of Civil Engineers frames it well: the goal is not to slow AI use but to design workflows and review processes so engineers can genuinely evaluate outputs, preserving learning and accountability rather than blindly accepting what the machine says.
CECA's May 2025 report notes AI tools are already entering businesses across the sector, sometimes unwittingly, and sets out recommendations on managing risk, cross-industry collaboration and upskilling the workforce.
What government and regulators add
The government's guidance for its own departments, originally the Cabinet Office generative AI framework, now superseded by the AI Playbook for the UK Government, sets a red line worth adopting in any construction business: never enter official or sensitive information into public generative AI tools unless it is already public, and build safeguards and human review into higher-impact uses. The NCSC's guidance helps leaders ask informed security questions, and the ICO sets out how UK GDPR applies when AI touches personal data: accountability, transparency, fairness, accuracy and security.
BSI's governance guidance pulls it together: AI governance is a lifecycle discipline covering data governance, model lifecycle management, risk and compliance, and ongoing monitoring, not a one-off policy document.
What this means in practice
Reading the guidance side by side, a consistent playbook emerges for a construction business of any size:
- Set data red lines first. Decide what can never go into public tools. If data cannot leave your environment, use private deployments instead.
- Keep a named human accountable for every material output. AI drafts, your people decide. No automatic issue of instructions, notices or compliance decisions.
- Check proportionately. Review material outputs properly; dip-sample high-volume automated ones. Blanket 100 per cent checking is not required, and pretending to do it helps nobody.
- Write it down. A short risk register and a record of what you rely on each tool for is most of what good governance requires.
- Train by role. The risks a bid writer runs are different from the risks a QS or an H&S manager runs. Generic awareness training misses both.
The point
The built environment does not need a blanket ban on AI tools. It needs clear rules on data, responsibility and checking, so the technology supports professionals rather than quietly bypassing governance.
Usage is already rising sharply. The practical question for your business is not whether AI gets used, but whether it gets used under your rules or under nobody's.