AI in Project Management Statistics: Adoption Rates, Industry Trends, and Key Findings

AI adoption in project management has moved beyond isolated experiments, but the headline percentage depends on what a survey counts as adoption. Some organizations use AI for occasional drafting or meeting summaries; others embed it in planning, forecasting, reporting, and decision support. Those are different levels of operational maturity, and they should not be treated as equivalent.
Recent surveys point in the same broad direction: use is widespread and growing, while consistent integration remains harder. The most useful statistics therefore reveal not only how many organizations report using AI, but also where it is being applied, what obstacles remain, and whether adoption is producing measurable delivery improvements.
AI adoption rates: what the latest surveys show
The Association for Project Management (APM) reported in September 2025 that 70% of 1,000 UK project professionals said their organization was using AI, compared with 36% in its 2023 survey. A further 29% said their organization planned to adopt it, leaving just 1% reporting neither current use nor plans. These figures indicate rapid diffusion, but they measure reported organizational use, not the proportion of projects managed primarily by AI.
Project Management Solutions' State of Project Management 2026 research found that 74% of organizations use AI-supported practices to help meet their goals. The same report found that 82% expect AI to have a great or very great impact on project management by 2030. Because the surveys use different samples and definitions, the 70% and 74% results should be read as converging evidence of broad adoption, not as a precise year-on-year increase.
The distinction between access and integration matters. An employee using a general-purpose assistant to summarize notes counts as AI use in many surveys. A project environment in which approved tools draw on governed project data, flag schedule risk, and feed decisions into established controls represents a more mature capability. Adoption statistics rarely separate these states cleanly.
Industry trends: adoption is spreading beyond technology
APM's 2025 sector comparisons show AI use rising across several industries. Construction moved from 15% reporting current use in 2023 to 75% in 2025; engineering rose from 35% to 73%; financial services increased from 23% to 66%; and transport reached 71% in 2025. These are survey responses about organizations using AI in projects, not audited counts of deployed AI systems.
The pattern challenges the assumption that project AI is mainly a software-sector phenomenon. Construction and engineering teams can use it to support document review, schedule analysis, risk registers, and change reporting. Financial services teams may focus on governance, regulatory documentation, portfolio reporting, and analysis of dependencies. The viable use cases depend on data quality, process maturity, and the consequences of an incorrect recommendation.
Architecture and construction data reinforces the trend while illustrating why survey populations matter. The Royal Institute of British Architects' 2025 AI report found that 59% of practices used AI at least occasionally on projects, up from 41% in 2024. Adoption was reported by 83% of large practices, 64% of medium-sized practices, and 48% of small practices. This is evidence about architecture practices, not the entire construction industry, so it should be used as a related indicator rather than merged with APM's figures.
Where project professionals are using AI
APM's 2025 research identified task and schedule automation, resource allocation, risk analysis and forecasting, and reporting and dashboarding among the functions respondents said had benefited most. Each was selected by around half of AI users who had seen a benefit: 50% for task and schedule automation, 50% for resource allocation, 50% for risk analysis and forecasting, and 49% for reporting and dashboards. Stakeholder communications followed at 43%. These figures describe perceived areas of benefit, not independently measured productivity gains.
Project Management Solutions' 2026 report similarly rated automating repetitive tasks highest among its listed AI practices, at 3.7 on a five-point scale. Improving productivity scored 3.6, analyzing large amounts of data 3.5, and automating reporting 3.4. Optimizing project schedules and improving budgeting accuracy scored lower, at 2.6 each. The gap suggests that teams are more comfortable using AI to accelerate repeatable work than to rely on it for core planning judgments.
For project managers, this points to a sensible adoption sequence. Start with tasks where outputs can be checked quickly and errors are recoverable, such as meeting summaries, first-draft status reports, document classification, and initial risk descriptions. Move toward forecasting and decision support only when the organization can validate input data, compare predictions with actual outcomes, and define who has authority to act.
Key statistics and how to interpret them
The figures below bring together recent surveys with different populations and definitions. They are useful for identifying direction and priorities, but should not be combined into a single global adoption rate.
Measure | Finding | What it means |
UK project professionals reporting organizational AI use | 70% in APM 2025, up from 36% in 2023 | Reported adoption has expanded quickly. |
Organizations using AI-supported practices | 74% in Project Management Solutions 2026 | AI use is common, but depth of integration varies. |
Practices using AI at least occasionally on projects | 59% of architecture practices in RIBA 2025, up from 41% in 2024 | Adoption differs by sector and firm size. |
AI fully embedded in workflows | 27% of project professionals in APM 2026 | Embedding AI broadly remains a separate milestone from trying tools. |
Organizations citing workflow integration as an AI challenge | 63% in Project Management Solutions 2026 | Implementation, not just tool availability, is a major barrier. |
Organizations expecting major impact by 2030 | 82% in Project Management Solutions 2026 | Expectations are high, but they are not evidence of realized benefits. |
Three conclusions stand out. First, adoption is no longer limited to a small group of early adopters. Second, occasional use and full workflow integration are different states, so organizations should measure them separately. Third, implementation friction remains substantial: Project Management Solutions found that 63% cited integrating AI into existing workflows as a challenge, ahead of resistance to change at 30% and skepticism about value at 23%.
Why adoption does not automatically produce better project outcomes
A high adoption rate does not prove that projects finish faster, cost less, or achieve more benefits. Many surveys ask whether AI is used or whether respondents perceive a benefit. Those measures are useful, but they do not establish causation. Teams that already have strong data, capable leadership, and disciplined delivery practices may be both more likely to adopt AI and more likely to perform well.
The most common failure mode is adding a tool without redesigning the work around it. AI-generated forecasts can create false confidence when schedules are incomplete, dependencies are outdated, or resource estimates are politically optimistic. Generated status reports can make weak source data look authoritative. A model may identify a pattern, but a project manager still needs to test whether the explanation fits the delivery context.
Governance should therefore match the risk of the use case. Define which data can be entered into approved tools, who reviews generated outputs, how material errors are escalated, and which decisions must remain with accountable people. For forecasting, record predictions and compare them with actual results over time. For reporting, check factual claims against the project system of record before distribution.
Skills are another constraint. APM's 2025 survey found that 49% of respondents using AI cited technical knowledge and training as a challenge, 44% cited security or data privacy, 42% cited workflow integration, and 41% raised accuracy or trustworthiness concerns. Training should therefore cover both prompt use and professional judgment: checking outputs, recognizing unsupported claims, protecting sensitive data, and knowing when not to use AI.
How to measure AI adoption in a project organization
A useful measurement framework separates activity, integration, and outcomes. Activity metrics show whether people use the tools, such as active users, frequency of use, or the share of projects with an approved AI use case. Integration metrics show whether AI is part of controlled delivery processes, such as the percentage of AI-assisted reports reviewed, forecasts linked to current project data, or workflows with documented ownership.
Outcome metrics test whether the change is worth keeping. Compare reporting cycle time, forecast error, schedule variance, rework, issue detection lead time, and stakeholder satisfaction before and after implementation. Where possible, compare similar projects or introduce the capability in stages. Record changes in scope, staffing, project complexity, and other factors that could explain the result.
Do not treat the number of prompts, licenses, or AI-generated documents as proof of value. A tool can be popular while creating rework or adding another approval step. A credible business case connects a defined use case to a baseline, a measurable target, an accountable owner, and a review date. If results do not improve, change the workflow or stop using the feature.
FAQs
What percentage of project managers use AI?
There is no single global rate because surveys measure different populations and definitions. APM reported 70% organizational use among UK project professionals in 2025, while other research measures AI-supported practices or occasional project use.
Which project management activities benefit most from AI?
Recent survey responses highlight task and schedule automation, resource allocation, risk analysis, forecasting, and reporting. Benefits depend on reliable data and human review, and reported perceptions should not be mistaken for independently verified productivity gains.
Will AI replace project managers?
The evidence is stronger for task assistance than wholesale role replacement. AI can draft, summarize, classify, and analyze, but accountability, negotiation, stakeholder trust, trade-off decisions, and response to ambiguity still require human judgment.
Conclusion: AI in Project Management Statistics: Adoption Rates, Industry Trends, and Key Findings
Recent research shows that AI use in project management is widespread and expanding across industries. Yet adoption percentages tell only part of the story: occasional experimentation, targeted automation, and governed workflow integration are materially different capabilities. The next measure of progress should be whether AI improves delivery outcomes, not simply whether an organization has introduced it.
Over the next two years, a reasonable expectation is that organizations will focus less on proving that AI can generate content and more on integrating approved tools with project data, reporting, and risk controls. This is a forecast, not a guaranteed outcome. Progress will depend on data quality, security requirements, staff capability, and evidence that AI improves decisions without creating unacceptable new risks.




































