Project and programme management has developed sophisticated methods for planning, controlling and reporting delivery. Yet much of the information used to manage complex programmes is still collected and reviewed periodically.
Progress reports are assembled, schedules updated, risks reviewed and management information produced at agreed intervals.
These activities provide necessary controls, but they often create a delay between something happening and the programme recognising its significance.
AI creates the opportunity to move towards a more continuous understanding of delivery.
Information already exists across schedules, supplier communications, technical records, design approvals, testing results, commercial systems and everyday project discussions.
AI could help interpret and connect these sources, identifying changes that affect delivery without requiring every observation to be manually transferred into a reporting tool.
Consider a delayed design approval. Its consequences may extend beyond the design activity to procurement, installation, commissioning and operational readiness.
A continuously maintained information model could identify these relationships, assess potential schedule consequences and bring the issue to the attention of the responsible delivery team.
The distinction is between reporting a problem after it has affected progress and identifying its likely consequences while there is still an opportunity to intervene.
This also has implications for risk management.
Traditional risk registers are frequently reviewed on a periodic basis, often separated from the daily activities that create or resolve the risks.
AI could support a more continuous approach, recognising emerging risks from actual delivery information, tracking their relationship to programme activities and assessing whether mitigation actions are having the intended effect.
Cost and schedule forecasting could similarly draw on a richer understanding of progress, dependencies and outstanding work rather than relying primarily on periodic estimates.
None of this removes the need for competent programme leadership.
AI-generated assessments depend on the completeness and reliability of their source information. Forecasts need assumptions, uncertainty and evidence. Decisions involving safety, contractual responsibility, cost or operational consequences require appropriate human authority.
The role of programme management could increasingly shift away from assembling information towards interpreting evidence, challenging forecasts and making timely decisions.
Formal governance and contractual reporting would remain necessary, but could be informed by a continuously updated picture of actual delivery.
The most important opportunity may be to move project controls from a periodic reporting activity towards an integral part of everyday delivery.