FCA authorised · FRN 724952 0117 325 0027Quote & buy →
Aipedia: Building services engineers
AI fault detection can find faults in building services faster than manual checks, but engineers who configure, interpret or sell these services take on new advisory and product risks. The line between professional advice and a software product affects which insurance responds.
Part of: AI in building services engineering
In short
Building services consultants increasingly use AI to support commissioning, analyse trend logs and run ongoing fault detection and diagnostics after handover. Risks include missed faults, false reassurance, advice given long after practical completion and data or cyber exposures. CIBSE Code clause 4.2, the Statement of Ethical Principles (Principle 5 on autonomous technologies) and Part 2A duties during construction apply. PI usually covers negligent professional advice, but if the firm supplies a software platform, claims may be argued as product failures, which PI wordings may treat differently. Check how your policy describes your services.
During commissioning, AI tools compare trend data with design intent, flag sensors reading out of range, identify valves or dampers not responding and generate snag lists. After handover, fault detection and diagnostics services monitor plant continuously, rank faults by energy or comfort impact and recommend fixes. Some consultancies sell this as a subscription, bundling analytics software with engineering review.
The attraction is clear: a single analyst can watch many buildings, and faults that once went unnoticed for months are flagged within days. But the value depends on data quality, the rules or models the tool uses and how quickly people act on alerts. A consultancy offering the service is often giving advice about a building it has never visited, based entirely on what the data shows.
AI fault detection can miss faults it was not trained to recognise, miss faults hidden by faulty sensors, or rank a serious fault low because its energy impact is small. Clients may assume that a system with no alerts has no problems. Commissioning reports drafted by AI may describe systems as tested when data was incomplete.
Illustrative scenario: a consultancy provides AI fault detection for a school's heating and ventilation under a monthly service. A failed humidity sensor causes the system to suppress alerts for a ventilation unit. Months later, poor air quality and condensation damage are found. The client claims for remedial works and alleges the service should have detected the problem, while the consultancy argues the software performed as described.
This question shapes the insurance. If the consultancy's engineers review the data and give advice, a claim is likely to be framed as negligent professional advice. If the client bought access to a software platform, the claim may be framed as a product that did not work as described, or a breach of contract terms in a licence. PI policies are written around declared professional services, and some contain exclusions for software or products supplied. Firms that build or resell analytics should make sure their activities are described accurately to insurers and consider whether technology-related cover is needed.
Claims for negligent commissioning advice or fault analysis usually fall under PI, subject to the wording, including claims made after handover where the service continues. Cyber incidents affecting the monitoring platform usually belong with cyber cover; see cyber insurance for engineers. Our page for building services consultants explains how consultancies' different activities are insured.
Related: AI in BMS, AI and O&M manuals 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 commissioning and ongoing monitoring services.
It can be either, depending on what is sold. Engineering review and advice usually count as professional services, while access to a software platform may be treated as a product. The distinction affects how PI responds.
It should not on its own. AI can support commissioning by analysing data, but a competent engineer should confirm systems perform as intended, and reports should not overstate what was tested.
Usually, if the advice is part of your declared professional services and the policy is in force when the claim is made, subject to the wording. PI is normally written on a claims-made basis.
A cyber incident affecting your systems or data usually sits with cyber insurance, and some PI policies contain cyber exclusions. Check both policies together so there is no gap.
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.
