The Time for AI Governance Is Now
By: Keisha Lewis, PMP
As we all know by now, artificial intelligence has begun disrupting the way we manage projects, and it will continue to do so even more in 2024 and beyond. This is not a bad thing; it’s exciting! As project professionals, we're lucky to be more than just observers, but active participants in this transformation. I am an advocate and cheerleader for the integration of AI in project management. But I’m also cautious, and I know that the use of AI needs to be managed and controlled.
AI is automating fundamental project management activities, from risk management and scheduling to meetings and stakeholder communication. According to a Gartner study, 80% of today's project management tasks are expected to be taken over by AI by the year 2030.
Scary? Well, instead, picture this: a future where project managers no longer need to spend countless hours on data collection, tracking and reporting because AI has got it covered. A future where mundane aspects of project management are things of the past, giving us more time to focus on the strategic and human elements of our roles. Wouldn't that be great?
Scary or exciting or both, we should all be able to agree on one thing: we must adopt a structured approach to manage AI in our project work. And that's where the concept of AI governance comes into play.
What is AI Governance?
Let's define what we mean by AI governance, specifically in the context of project management. AI governance for project management is a structured approach that standardizes and formalizes the oversight an organization applies to the usage of AI in their project management functions.
But it's not just about laying down the rules; it is a comprehensive framework designed to direct, manage, and monitor the AI-powered project management activities. This approach ensures the responsible and ethical use of AI and mitigates risks.
AI governance is also about establishing accountability. It’s about making sure that someone is responsible for every decision made and every action taken by the AI systems.
Finally, AI governance is a critical part of earning and maintaining the trust of your project stakeholders.
Consequences of a Lack of Governance
Let's now discuss the dangers that you may face when there is no AI governance plan for project management. I want you to consider a world where you've allowed AI to integrate into your project management processes and tools without any oversight or direction. And some of you don’t have to imagine this, because it’s already happening within your organization – ouch!
Some of the consequences are:
Loss of Trust: Without clear governance, stakeholders might lose trust in AI-driven decisions if they feel they aren't transparent or accountable.
Increased Risk: Without a governance framework, there's a heightened risk of security breaches, misuse of data, or unintentional ethical violations. For more on data governance, risk and quality, check out PMI's Data Landscape of GenAI for Project Managers course.
Resistance to Adoption: A lack of clear governance can lead to uncertainty or fear among team members, resulting in a reluctance to adopt AI-driven tools and processes.
Poor Decision Making: Without proper governance, it's easy to lose control over AI applications. They can make decisions that humans wouldn't, leading to outcomes that deviate from your project's goals.
Non-compliance with Regulations: There is a growing body of laws and regulations around the use of AI. Without proper governance, you risk non-compliance, which could result in penalties, legal issues, and damage to your organization's reputation.
Recommended by LinkedIn
Decentralized vs Centralized Approach
The choice between a decentralized and centralized approach to AI governance in project management can impact effectiveness. Each offers opportunities and risks.
In a decentralized approach, each department or project team within the organization is responsible for its own AI usage. These units have the freedom to choose their own AI tools, methodologies, and data used to train the AI, depending on their specific project needs. While this approach allows for flexibility and customization, it could also lead to inconsistency in AI usage across the organization.
For example, in Company A, one team might choose to use a specific AI tool for risk management, while another team might prefer a different tool for the same task. The problem is that each tool may take a totally different approach, which leads to inconsistencies. Likewise, the data used for training the AI can vary from team to team.
So, while this approach allows each team to tailor their AI usage to their specific needs, it can lead to inconsistencies in project outcomes and stakeholder expectations.
On the other hand, a centralized approach is led by a single, enterprise-wide entity that makes decisions and sets standards for AI usage. This entity may be a PMO, Center of Excellence or other project management leadership group within your organization. This body manages the rules, methodologies, and approves AI PM tools for use, ensuring consistency across the organization. It also bears the responsibility for risk management, AI training, and compliance with regulations.
For example, Company B has a PMO that establishes a centralized AI governance system. The PMO vets AI-powered tools and vendors and determines which ones are approved for usage along with the acceptable use cases. So, let's say you find an AI tool that you'd like to use that automates scheduling. You would have to submit your request through a PMO intake process.
During this intake process, the PMO would evaluate the vendor and the tool based on a set of criteria. If the tool is approved, the PMO would define acceptable use cases, provide training and ensure the tool remains compliant with external regulations. All project teams within the organization must abide by these guidelines, creating uniformity and consistency across the company.
As you can see, both approaches come with advantages and drawbacks. The decentralized approach offers flexibility, speed, and the possibility of local optimization, but it can also lead to inconsistencies, duplication of efforts, and challenges in risk management.
On the other hand, a centralized approach fosters standardization, efficient use of resources, and comprehensive risk management, but this may slow down implementation and limit flexibility.
While both approaches have their merits, a centralized approach offers significant benefits with less risk.
Getting started
As you start your journey to AI governance for project management, you will typically find yourself falling within one of four scenarios:
Wherever you might be on your AI journey, the time for AI governance is now. On The Official PMI Blog, this article includes a free download to help you get started: a 27-page AI Governance template that covers governance principles, roles and responsibilities, readiness and maturity assessments, monitoring, risk management, data, alignment and much more.
This article was originally published on The Official PMI Blog.
This is where we step in – we help organizations hit the ground running with their AI Governance policies without having to struggle to understand the fast-changing AI landscape.
CEO, Flex PM Services | Subscribe "Next-Level Skills"
4moVery well put together Keisha Lewis, PMP! Thanks for sharing. I believe AI will supplement how we operate as Project Managers in things like reports generation, risk management, help creating project artefacts and other tasks that take too much of PM’s time away from their real work which is building supportive and meaningful connections with various stakeholders, which is the ultimate decider for the success or failure of any project (given other things are in order). You rightly pointed out that AI can help us immensely as long as a proper and solid governance is not only in place but very well executed too. Reading through some of the valuable comments everyone had shared earlier, I do feel as well, this governance needs to be dealt with due importance and NOW! As a project manager, I have already started learning about how GenAI works, so that I am better prepared to deal with it. And that needs to be done along side setting a governance up as well I think. Project Management Institute
Internet Researcher about malnutrition disease also #healthcare & yoga; Stock trading experience; Online clothing imports; Desktop Publisher #aigovernance & Personal friend of Queen Elizabeth.
5moWho didnt publish my comment?
Transformation Leadership | Complex Program Management | Strategy | PMO | Management Consulting
5moGood article. If, as postured, 80% of project management tasks will be handled using AI by 2030, I can continue to see the real value being in the other 20%... Most definitely, there are a myriad of tasks involving data collation and synthesis and AI offers enormous opportunities to do these tasks quicker and more comprehensively than today. The challenge will continue to be more about how to use the data to make effective decisions more than simply having the data. I can see the immediate governance challenge being as much about what data do we need to make effective decisions regarding projects in our organisation as how do we gather that data. And most certainly, it would appear that this needs to be managed consistently across the enterprise portfolio of projects and programs.