Project Management Institute has released what it describes as the first published global standard for applying artificial intelligence in portfolio, programme and project management, as organisations race to adopt AI faster than governance frameworks can keep pace.
Project Management Institute (PMI) has announced the release of The Standard for Artificial Intelligence in Portfolio, Program, and Project Management, a new global standard designed to guide the responsible use of AI across professional project work.
The publication marks a significant moment for the project profession, with PMI positioning the standard as the first of its kind to address how artificial intelligence should be designed, governed, deployed and overseen within portfolios, programmes and projects.
The release comes as businesses and public sector organisations accelerate investment in AI systems, AI-enabled products and AI-augmented workflows. Yet while much of the global debate has focused on regulation, ethics, technology capability and organisational risk, PMI argues that less attention has been paid to the delivery environment through which AI transformation actually happens.
In most organisations, AI adoption is not delivered in the abstract. It is delivered through projects. Whether an organisation is implementing a generative AI tool, automating internal workflows, embedding AI into customer services or building AI-led products, the work is typically scoped, funded, governed and delivered by project, programme and portfolio teams.
PMI’s new standard seeks to address that gap by giving project professionals a common framework for managing AI initiatives in a way that is practical, responsible and aligned with emerging governance expectations.
“Most conversations about AI governance focus on what the technology can do, and it’s often mind-blowing. But far less attention has been paid to how AI is actually delivered, which is almost always through projects,” said Pierre Le Manh, President and CEO of PMI. “AI transformation succeeds or fails in the projects and programmes that deliver it. This standard is about what makes AI deliverable at scale.”
The standard sets out eight guiding principles, five performance domains and a complete lifecycle framework for AI initiatives. According to PMI, it is technology-agnostic, meaning the guidance is intended to remain relevant across different tools, models and platforms as the AI market continues to evolve.
A central feature of the standard is its emphasis on human-in-the-loop oversight. Rather than treating AI deployment as a purely technical exercise, PMI’s approach places human judgement, governance and accountability at the centre of the delivery process. That principle reflects growing concern among regulators, business leaders and civil society groups that AI adoption must be transparent, explainable and subject to meaningful control.
The standard arrives during a period of intense international scrutiny over AI governance. Governments and regulators are developing risk-based approaches to AI oversight, while organisations are trying to understand how new requirements will affect investment, compliance, assurance and delivery. The European Union’s AI Act has become one of the most closely watched regulatory frameworks, while standards such as ISO 42001 are shaping expectations around AI management systems.
For project leaders, those developments create a practical challenge. AI initiatives increasingly require alignment between legal, audit, finance, technology, procurement, security, risk and business teams. Without a shared delivery framework, organisations risk creating fragmented governance arrangements that slow progress, weaken accountability or expose projects to avoidable risk.
PMI said its standard is designed to help project professionals navigate those decisions, from the earliest business case through to tool selection, risk management, ethics oversight, development, deployment and ongoing monitoring.
That focus is important because many AI failures are unlikely to stem from the technology alone. Poorly defined objectives, weak sponsorship, inadequate stakeholder engagement, flawed data assumptions, unclear accountability and insufficient change management can all undermine AI initiatives. These are familiar project delivery risks, but AI introduces additional complexity around bias, transparency, data quality, model behaviour, regulatory exposure and public trust.
By framing AI adoption through portfolio, programme and project management, PMI is making the case that responsible AI is not simply a matter for data scientists, technology vendors or compliance teams. It is also a delivery discipline.
The new standard provides project professionals with a structure for asking sharper questions at each stage of AI work. What business problem is the AI initiative solving? How will benefits be measured? What risks are specific to the technology, data and operating context? Who remains accountable for decisions supported or produced by AI? How will affected stakeholders be consulted? What controls are needed before, during and after deployment?
These questions are becoming more urgent as AI moves from experimentation into enterprise-wide implementation. Many organisations have already moved beyond pilot projects and are now exploring how AI can reshape operations, improve productivity, support decision-making and create new products or services. The opportunity is considerable, but so is the delivery burden.
For portfolio leaders, AI investment raises questions about prioritisation, value, risk appetite and organisational readiness. For programme managers, it creates the need to co-ordinate multiple workstreams across technology, policy, people, process and assurance. For project managers, it demands disciplined delivery, clear governance and the ability to translate technical ambition into practical outcomes.
PMI said the standard was developed with input from its practitioner community, drawing on real-world experience from professionals working across project environments. That practitioner-led approach is intended to ensure the standard is not merely theoretical, but useful for the decisions project teams face as AI becomes embedded in everyday delivery.
Dr Kelly Heuer, Vice President of Learning at PMI, said the standard gives project professionals a common language for engaging with senior stakeholders and cross-functional teams.
“Project professionals need a common operating language that helps them align legal, audit, finance, technology and business teams around how AI work is approved, governed and delivered,” said Dr Heuer. “This standard gives project leaders running AI initiatives the tools to drive ethical implementation and the language to defend their work upstream, increasing the likelihood of project success.”
The release also reflects the changing role of the project profession. As AI adoption becomes a board-level priority, project professionals are increasingly expected to operate not only as delivery managers, but as integrators of strategy, governance, risk and change. The success of AI transformation will depend heavily on their ability to bring structure to fast-moving and uncertain environments.
That may prove one of the standard’s most important contributions. AI can often be discussed in sweeping terms, with promises of disruption, automation and productivity gains. But organisations do not realise value from AI through ambition alone. They realise it through well-governed delivery.
A global standard gives project professionals a clearer basis for challenging weak assumptions, strengthening oversight and ensuring responsible implementation is built into the work from the start, rather than added as an afterthought.
The standard is available immediately to PMI members worldwide as a free digital download. Print editions are also available for purchase. Non-members can access the digital edition for $74.95 or through PMI membership.
For organisations investing heavily in AI, the publication offers a timely reminder that successful transformation depends on more than technology selection. It depends on governance, accountability, delivery discipline and human judgement.
As AI becomes a defining feature of modern organisational change, PMI’s message is clear: the future of AI will not only be shaped by what the technology can do, but by how well it is delivered.
