FCA authorised · FRN 724952 0117 325 0027Quote & buy →
Aipedia: Building services engineers
AI tools can size and select heat pumps quickly, but a design based on wrong heat loss, flow temperatures or emitter sizes still belongs to the designer. Heat pump complaints often focus on comfort and running costs, which clients may treat as promises.
Part of: AI in building services engineering
In short
Designers use AI to estimate heat loss, select heat pumps, size emitters and cylinders and predict running costs for homes and commercial buildings. Errors usually come from inputs or optimistic assumptions about flow temperatures and efficiency. Undersized systems leave rooms cold, and oversized or poorly controlled systems cost more to run than predicted. CIBSE Code clauses 4.2 and 1.3 cover competence and misleading statements, and Part 2A regulation 11K applies where the work is notifiable. PI usually responds to negligent design claims, but running-cost guarantees may fall outside cover.
AI tools estimate heat loss from floor plans, photos or energy certificates, suggest heat pump capacity, select emitters for a chosen flow temperature, size hot water storage and generate running-cost and carbon comparisons. Larger schemes use AI to model cascaded or hybrid systems and to suggest control strategies.
These outputs often go straight into quotes or client reports, sometimes produced by a consultant for an installer or a homeowner, so the gap between an AI suggestion and a client-facing promise can be very short.
Any certification scheme or grant requirements that apply to the project sit on top of these duties.
AI heat loss estimates from limited data can be well off for older homes with uncertain fabric. Tools may assume flow temperatures the existing radiators cannot deliver, or efficiencies the system will not achieve in practice. Running-cost figures may use tariffs or usage patterns that do not match the household.
Illustrative scenario: a consultant uses an AI tool to design a heat pump system for a terraced house, relying on an estimated heat loss and keeping existing radiators at a low flow temperature. The report includes an annual running-cost figure. In the first winter, two rooms never reach temperature and bills are well above the figure. The homeowner complains and claims for new radiators, redecoration and the difference in bills, saying the report was a promise.
On commercial buildings the risks change. AI tools may propose cascaded heat pumps, hybrid systems or heat pumps serving domestic hot water and space heating together, and the control strategy becomes as important as the plant size. Errors can show up as plant cycling, poor hot water recovery, high electrical demand or a need for an upgraded incoming supply that nobody budgeted for. The engineer should check that electrical capacity, plant space, acoustic limits and refrigerant considerations have all been considered, not just the heating duty the tool optimised for. Where a heat pump replaces gas plant in an existing building, the existing distribution system usually needs its own assessment.
A claim that a system was negligently designed and does not heat the home is usually within the scope of PI, subject to the wording. Claims based on a guaranteed running cost or a performance warranty are more difficult, because policies may exclude or limit liability assumed under contract beyond reasonable skill and care. See fitness for purpose versus reasonable skill and care and the M&E specification claim scenario. If you also install or commission systems, our page on PI for heating engineers explains how design and installation risks are insured differently.
Related: AI load calculations, AI net zero advice and MEP PI and AI errors. Back to AI in building services engineering.
Apex arranges PI for building services engineers. A broker reviews your cover and how your policy treats AI-assisted heat pump design and any performance statements in your reports.
You can share estimates, but present them as estimates with the tariffs and assumptions used. CIBSE Code clause 1.3 prohibits statements known to be untrue or misleading, and figures read as promises create avoidable claims.
Usually because of inputs: estimated heat loss, assumed fabric, or a flow temperature the existing emitters cannot deliver. The calculation may be sound while the assumptions are not.
PI usually responds where the claim alleges negligent design, subject to the wording. Guarantees of comfort or running cost can go beyond reasonable skill and care, which policies may not fully cover.
Often yes. Design advice is usually insured under PI, while installation work and physical damage tend to sit with public and products liability. A broker can check how the two fit together.
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 callbackApex 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.
