Aipedia: Principal designers and building safety

Is AI use a principal designer competence question?

Aipedia, by the Apex broking team · Last reviewed 2026-10-27

Relying on AI can become a competence question when the principal designer cannot judge or verify what the tool produces. Part 2A makes competence a legal duty, so the issue is how AI is used, not whether it is allowed.

Part of: AI in principal design and building safety

In short

Part 2A of the Building Regulations requires principal designers to have the skills, knowledge, experience and behaviours, or organisational capability, for the role (regs 11F and 11G), and to notify the client if they cease to do so (reg 11I). The CROSS-UK panel in Report 1554 cited these duties against unverified AI design outputs. Using AI is not itself incompetence, but accepting outputs no one in the firm can check may be. Competence findings can affect Gateway 2 approvals, civil claims and, through building liability orders, associated companies.

Why competence is the right lens

Part 2A of the Building Regulations does not mention AI. It does make competence a legal duty. If a principal designer relies on AI in a way a competent PD would not, the question is not whether AI was allowed but whether the PD met its competence duties. The AI for principal designers overview sets this in context.

The competence duties

Part 2A reg 11F: competence means the skills, knowledge, experience and behaviours (individuals) or organisational capability (organisations) needed for the work.
Reg 11G: the PD must have the competence needed for the PD role. An organisation PD must designate an individual to manage its PD functions after taking reasonable steps to confirm that person is suitable.
Reg 11I: a PD that ceases to meet the competence requirements must notify the client.
Reg 11E: the client must check the PD's competence before appointment, and on HRB work ask about serious sanctions in the previous five years.

The CROSS-UK Report 1554 panel cited regs 11F, 11G and 11H against unverified AI design outputs, describing them as mandatory competence requirements for all dutyholders.

When AI use could become a competence issue

Using AI is not, of itself, a sign of incompetence. Possible concerns include:

skills and knowledge: the designated individual cannot judge whether an AI output is right, for example because it covers fire or structural matters outside their expertise;
behaviours: outputs are accepted without review because they look professional, which the CROSS panel warned about;
organisational capability: the firm has no process for verifying AI outputs, or the capability depends on a tool or person it no longer has.

The Engineering Council and RAEng Statement of Ethical Principles asks engineers to consider AI risks and maintain competence in novel areas. RICS members acting as PD must follow the RICS AI standard. Where a PD's capability has genuinely fallen below what the role needs, reg 11I notification may become relevant. That is a judgement for the PD on the facts.

How competence findings can spread

Competence questions rarely stay in one place:

on HRBs, the client signs a competence declaration with its Gateway 2 application, so doubts about the PD's competence can affect approval (see AI in Gateway 2 applications);
the BSR is the building control authority for HRBs, and competence evidence is part of what it sees;
in a civil claim, a finding that reliance on AI fell below reasonable competence supports negligence;
under section 130 of the Building Safety Act 2022 the High Court can extend liabilities under the Defective Premises Act, section 38 of the Building Act 1984, or arising from building safety risks, to associated companies where just and equitable.

Apex's guide to building liability orders and note on the reasonable competence standard explain these in more depth.

How PI usually responds

PI usually responds to negligence claims, including those where AI reliance is alleged to have fallen below reasonable competence, subject to the wording. Regulatory sanctions and fines are generally not covered. If a PD knows of circumstances that could lead to a claim, such as a competence query from a client or regulator, notifying promptly is usually a policy condition. See PD insurance and AI.

Practical steps

Map which AI tools the PD function uses and who is competent to check each output.
Keep the designated individual's competence evidence current, including AI.
Set a verification step for every AI output that feeds a statutory document.
Know when to bring in a specialist rather than rely on a tool.
Review the guide to PD duties and AI for how each regulation applies.

How Apex helps

Apex arranges PI for principal designers. A broker can help you describe your competence controls for AI to insurers and explain when to notify circumstances.

Sources

Frequently asked

Is using AI a breach of reg 11G?

Not by itself. Reg 11G requires the PD to have the skills, knowledge, experience and behaviours, or organisational capability, for the role. The question is whether the PD can judge and verify what the AI produces.

When might reg 11I notification apply?

Reg 11I requires a PD that ceases to meet the competence requirements to notify the client. Whether reliance on AI, or loss of a key person or process, has that effect depends on the facts. It is a judgement the PD must make.

Can competence findings affect associated companies?

Possibly. Section 130 of the Building Safety Act 2022 lets the High Court extend certain building liabilities to associated companies where just and equitable. A negligence finding against a PD could form part of the underlying liability.

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.