FCA authorised · FRN 724952 0117 325 0027Quote & buy →
Aipedia: Project managers
AI can draft a programme or predict delay in seconds, but the project manager remains responsible for whether the sequence is buildable and the forecast is reasonable. Programmes are relied on by clients, funders and contractors, so a confident but wrong AI forecast can drive real financial decisions.
Part of: AI in project management
In short
Scheduling tools can generate draft programmes from a scope, flag likely delay from progress data and suggest recovery sequences. The risk is a programme that looks professional but ignores site constraints, lead times or the contract's own sequencing, or a delay forecast that is presented as more certain than it is. Clients and funders rely on these outputs to release money and plan occupation. For RICS-regulated PMs, a material AI output needs a written reliability decision. Programme errors are usually handled as negligence claims under PM PI, subject to the wording.
APM-funded research found planning was where AI was most used in project management, and that its value rose with complexity. Typical uses can include generating a first-cut programme from a scope or bill of quantities, checking logic and float, forecasting completion from progress returns, and modelling recovery options after a delay event. Some tools read the contractor's programme and comment on it, which is useful when a PM must accept or reject a submitted programme under the contract.
The weakness is the input. A model can only sequence what it is told about: if access routes, phasing for an occupied building, utility outages or long-lead items are missing from the data, the programme will look complete and still be wrong. Delay prediction tools also learn from past projects, which may not resemble yours.
The core duty is to use reasonable skill and care, measured against a competent PM, not against what the software produced. For PMs who are RICS members or in RICS-regulated firms, the RICS AI standard adds specific steps where an output has a material impact: a written decision on its reliability made by a named, qualified person (4.2) and written notice to the client about AI use (4.3). The RICS Rules of Conduct Rule 2 (competence) and Rule 3 (diligent service) apply to the finished programme. On construction projects, a PM acting for the client is helping the client meet its duty under CDM 2015 regulation 4 to make suitable arrangements, which includes allowing adequate time for each stage.
Illustrative scenario. A PM uses an AI scheduler to produce a programme for a refurbishment. The output assumes trades can overlap in a way the building's access arrangements do not allow, and it uses standard lead times for plant that, on this project, had long delivery periods. The programme goes into the funding pack. When the job overruns, the funder applies a contractual remedy and the client alleges that, had it been given a realistic programme, it would have negotiated different terms and a later handover date.
Other patterns include a delay forecast that wrongly attributes critical delay, weakening the client's position in a later delay claim, and an AI critique of a contractor's programme that misses a logic error the PM then accepts.
Under a PM policy the claim is usually about the PM's advice, so it is usually treated as alleged negligence regardless of the tool used, subject to the policy terms. Points that commonly matter are whether the programme was within the services described in the proposal, whether the appointment limited the PM's responsibility for contractor programmes, and whether there is any AI-related clause. Delay claims can be large and slow, so the limit and how defence costs sit against it deserve attention. There is no statutory PI requirement for PMs, but appointments usually set a minimum. See the project managers PI guide and does PI cover AI mistakes?
Apex arranges PI for project managers. A broker can review your cover, your appointment terms on programme responsibility and how your policy treats AI-assisted planning; see also AI cost forecasting and the AI in project management hub.
Usually the question is whether you exercised reasonable skill and care, not whether the software failed. If a competent PM would have spotted the error, the client is likely to look to you. Any claim against the software supplier is a separate matter that depends on its contract terms.
It depends on your appointment and any reliance letter or collateral warranty. Funders commonly take the programme into account when releasing funds, so check what reliance you have given before sending them an AI-assisted programme.
Keep the inputs you gave the tool, the version it produced, the changes you made and why, and who signed it off. If the programme is later criticised, that trail shows the judgement applied was yours, which is the basis of most PI defences.
RICS-regulated PMs must give written notice of AI use in advance. For others it depends on the appointment, but telling the client how the programme was built and what was checked by a person is good practice and helps if it is later challenged.
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.
