AI and the PMO: A Reality Check

When attempting to discuss Artificial Intelligence (AI) in the context of Project Management, lots of opinions and ideas crop up that provide insight into how jobs MAY change, what impact MAY be felt, and how it MAY change our working economy.
The latest Wellingtone Community Networking event took place to provide some real-world insight into AI, cutting through some of the overwhelming content that is out there.
To do that, we invited Dwr Cymru Welsh Water to join us as hosts and SMEs. Dwr Cymru Welsh Water has been through a real-world test of AI for the PMO, to write a Business Case for the wider organization, detailing the rules of engagement with AI to ensure a healthy and sustainable relationship with the technology.
Following the event I asked Microsoft Copilot to summarize the event with some key points of note from the conversation, questions, and answers throughout the event. And then I added the #HumanFirst nuances that AI is not yet able to gather.

• Importance of Context in AI Usage: Richard and Emma discussed the importance of providing context when using AI tools like Copilot. Richard emphasized that AI performs better when given specific contexts, such as organizational documents, to generate more relevant and useful outputs.
  • PMO reality: As with all PPM and other tools, data is king. Without having a good foundation of information sitting behind the solution, it simply will not have the required knowledge to deliver the insights we may be looking for.
  • PMO lesson: Make sure that you have all your background data, project documents, guidance, and results available for AI to interrogate – it’s not psychic.
• Using Copilot Across Different Platforms: Emma and Jade talked about using Copilot across various Microsoft platforms like Teams, PowerPoint, and Outlook. Jade shared their experience of using Copilot in Teams to expand on comments and provide risk mitigation suggestions.
  • PMO reality: As with all changes that bring behavior and cultural shifts, delivery teams need to ensure that the solution can be accessed and used in the most relatable way, so this requires education on using AI across all the MS (or another cloud) suite of applications.
  • PMO lesson: Don’t assume that people will just ‘get it’. Provide education and guidance on accessing AI in familiar settings, and allow an environment where people can fail fast, ask questions, and request support.
• Role of Copilot in Document Creation: Richard explained that Copilot shifts the user’s role from creator to editor, significantly reducing the time required to draft documents. This allows users to focus on refining and humanizing the content, making it more strategic and accessible.
  • PMO reality: This is an important message that PMO and other roles do not get deleted by AI, but instead provide a shift in focus from tactical gathering of data, to analyzing the insights to ensure that the right decisions can be made with information that is accessible, easy to understand, and relevant.
  • PMO lesson: Make AI part of the day job so that individuals are not afraid to use it. But make clear the rules of engagement. As an editor, it is your responsibility to ensure that the information you choose to share (regardless of its origin) is correct and appropriate for you to send. It does not and must not diminish your accountability for what you share with others.
• Enhancing PMO Communications: Emma highlighted how Copilot can improve PMO communications by handling the initial drafting of documents, thus allowing project managers to focus on making the content more engaging and effective.
  • PMO reality: Starting with a blank sheet of paper is always the most difficult part of creating anything. So using AI to create the bare bones of something, can make the creation process more efficient, and allows us to stay in the mindset of the PMO value chain.
  • PMO lesson: Use the PMO value chain to determine if everything generated and then refined by human hand has:
    • Responded to the perception of the value of the PMO Customer
    • Provided a way to measure that this new tool adds legitimate value to the user and the surrounding processes
    • Enabled the PMO and the project delivery team with an additional tool for communication
• AI in Decision-Making and Accountability: Nichola discussed how AI tools like Copilot can enhance decision-making processes by providing clarity and accuracy. This helps in maintaining a clear record of decisions and reduces the administrative burden on team members.
  • PMO reality: Given that one of the key challenges PMO teams face time and again is accountability of decisions made throughout projects, AI can enable a culture of accountability by providing a non-emotive record of decisions made, by whom, when, and their context.
  • PMO lesson: Encouraging AI ‘listening’ as part of the standard operation of meetings and boards, will provide the foundations for clearer accountability and remove the need for ‘minute taking’.
• Confidence and Accuracy in AI Outputs: Jade added that Copilot helps in validating actions and decisions during busy steering group meetings, freeing up time for more critical tasks. This boosts confidence in the accuracy of the outputs.
  • PMO reality: Decision-making and confidence of the individual are closely connected. Natural human behavior lends itself to the protection of oneself when a decision needs to be made. AI can help with the confidence of individuals as there is less human error in data analysis.
  • PMO lesson: AI can provide a platform for those who do not feel confident in being creators, as they can begin with the AI ‘starter for ten’, rather than starting from scratch. In this case, AI can be seen as a vehicle for improving capability in some areas for individuals.
• Future Dependence on AI Tools: Emma and Richard speculated on the future dependence on AI tools, comparing it to the reliance on calculators. They discussed how AI could become indispensable in project management, much like other technological advancements in the past.
  • PMO reality: Becoming dependent on new technology can have a big impact on organizations and the roles and responsibilities that sit within teams.
  • PMO lesson: Consider how teams are built in the organization, the hierarchy may need to change along with reporting lines.
• Ethical and Skill Implications of AI: Richard raised concerns about the potential skill atrophy due to over-reliance on AI. They emphasized the need for a mature organizational culture to support AI usage while ensuring that employees remain competent in their core skills.
  • PMO reality: Loss of skill is a real factor when considering AI in the longer term. Just like changing from manual to automatic, or self-parking cars. Eventually through use (or lack of practice), we will see skill erosion and this should be addressed from an HR perspective.
  • PMO lesson: Consider processes that will need to change such as how we recruit individuals (what are the questions we need to ask in the future), and individual performance development (how much does AI shape roles and education).

• Considerations such as:
  • Existing security policies
  • existing rules for sharing dataconfidentiality of projects and programs
  • Permission levels
  • Organizational AI strategy
• Understanding the real question that we are trying to answer with Artificial Intelligence
• The process that Dwr Cymru Welsh Water used
  • Test & Security Design
  • Proof of Concept
  • Pilot
  • Controlled Adoption
• Identifying the right group of people who can test, whilst remaining impartial to their findings to create the Business Case for the wider organization
• People need training and education across the whole spectrum of the following:

The results for Dwr Cymru Welsh Water include four main areas where AI provided value in the real world of PMO. Note that in the four sections, AI assists; it doesn’t replace and doesn’t negate the need for human intervention.

• Get ChatGPT to teach you how to talk to AI, improving the prompts you are using to get better results.
• Start tagging files early with intelligent keywords to enable AI to search effectively.

As our Wellingtone Community Networking events are a discussion, with representatives from around the globe there were some interesting questions asked and answered to give us a flavor of how organizational attitude impact the adoption of new technology like AI.We can see from the responses to our questions that all participants have either used AI or tried it but are not yet convinced whether it’s for them or their current role.

This tells us that there is lots of interest out there, as an industry we acknowledge the importance of AI and the fact that it will change our roles, but overall we are not yet sure how to apply it fully.
One of the key drivers for AI (whether organizations are using it or not) is the understanding that it will support decision-making in organizations – an important note seeing that many organizations struggle with accountability for decision-making at all levels.

Open questions from the network provided us with the challenges organizations are facing along with some of the anticipated benefits of implementing more AI driven processes.

Direct questions from the network included lots detailed in this article but maybe one of the most important because it will be asked by your PMO Customers is:

“If we still have to edit things, then what is the point of using AI?”

AI puts you in the editor’s seat as it is a) still learning and b) cannot make the decisions for us. Humans have to continue to be accountable for the information that they choose to share. AI can take a lot of the non-value work away from us so we can focus on the areas that enable successful delivery.

We can see from the test that Dwr Cymru Welsh Water has completed that Artificial Intelligence can provide an environment and culture for the development and growth of the project management practice as well as the individuals therein.

It will be important for organizations to provide a clear strategy for AI, a Business Case detailing what challenge or opportunity it will address, and security questions all answered before embarking on a full-scale deployment. But a small pilot within the PMO will bring the art of the possible can be brought to the forefront and open up the conversation.

To change culture, it takes time and it takes a shift in the conversation Sometimes those changes seem too small, but they do make a difference.

As long as we remember that AI will not replace humans (just yet), it just puts them in the editor seat, remaining responsible for sharing information in a trusted way to drive decision-making, then adoption can bring positive impacts to your organization.

Thank you to Dwr Cymru Welsh Water for sharing their journey with the Wellingtone Community.

Ready to get started on propelling your own PMO, with or without AI? Join our community to keep up with the latest in project management.