Why AI Capability Matters for Project Professionals
Most people running projects aren't project managers. They are finance leads, marketing managers, IT heads and operations people asked to manage a launch, rollout or restructure on top of their actual job. The project is the extra layer that lands when business-as-usual is already full, and the documentation is what gets done late on a Thursday.
That pressure helps explain why project delivery so often slips. One commonly cited statistic shows only 31% of projects finish on time, on budget and within scope (PM Study Circle, 2025). For many project leads, that will feel familiar. Most can recall a recent project where timing, budget or scope moved. There's usually too much to track and not enough time, and decisions end up made with incomplete information because pulling everything together takes longer than anyone admits.
AI is already showing up in project work, even if not always in a planned or consistent way. The number of organisations using AI on at least half of their projects jumped 86% between 2023 and 2024 (PMI, 2024). According to GetApp, 63% of project professionals believe AI could help with the main challenges they face (GetApp, 2024).
Many organisations already use project management tools with AI built in, including platforms like Asana and monday.com. These features can help, but they usually work best with information already inside the platform. A lot of project work still sits elsewhere: the stakeholder update drafted in a hurry, the meeting notes that never made it into the system, the risk conversation that happened over email.
That is why people often go straight to ChatGPT, ask it to generate a project update with little context and get something generic back. They try it once, decide the manual version is faster and close the tab. The issue is knowing how to brief AI, what to leave out and what needs a human check before a sponsor sees it. That capability is not built by learning a tool in theory. It comes from practising with familiar project tasks, using enough context and review to make the output useful.
Why The First Prompt Almost Always Falls Flat
The first prompt usually fails because it asks AI to guess too much. A project update needs context: who it is for, what has changed, what is blocked and what decision is needed. It's the same problem as asking a new hire to do important work without explaining the audience, the stakes, or what 'done' means in this situation.
Once you give it real project detail, the draft starts to improve. Think about who the update is for and what they care about, then build the prompt around that. Sponsors want to know what moved, what is blocked and what needs a decision. Delivery teams usually need owners, actions, dependencies and next steps.
Before pasting notes into any AI tool, always remove names and sensitive information, especially if you are not using an approved enterprise platform. The clearer the context, the more useful the draft becomes, and it usually comes back faster than starting from scratch.
This shows up most obviously after the meeting. Six people, ninety minutes, a whiteboard photo and a vague sense that someone agreed to 'look into it.' By 4pm, you’re scrolling back through Teams to work out whether anyone actually owns the task. Give AI a clear brief and the meeting notes, and you can have a usable first draft while the conversation is still fresh. In practice, that usually means turning rough notes, stakeholder comments and half-formed actions into something structured enough for the project lead to review, correct and send on.
Where AI Can Get Project Work Wrong
AI can produce something that looks credible enough to send. That's exactly how people get caught out.
Ask AI for a timeline and it will give you one. Unless you provide the missing context, it will not account for team capacity, undocumented dependencies or the stakeholders who are away in December. It'll produce something that looks plausible. It won't flag what it doesn't know.
It also cannot read the room. It will miss the subject matter expert who agrees in meetings but delivers late, or the vendor relationship that is more fragile than the contract suggests. That knowledge lives with you. AI handles the structural work well: drafting, organising, pulling a first pass together. Your judgement is still what matters when the decision depends on knowing the room.
Before anything goes to a sponsor or steering group, the project lead still needs to check the assumptions, fill the gaps and decide what is safe to send. AI earns its keep earlier, by turning rough notes into something the project lead can test. Treat it as a source of truth and the risk stays with you.
AI is still worth using, but a clean-looking update is not the same as a reliable one.
The Capability Gap: Knowing How to Use AI Well in Project Work
The IPMA AI Survey 2024 found that 69% of project management professionals name a lack of AI knowledge as their biggest barrier (IPMA Publications, 2024), ahead of budget, tools and everything else. The same survey found that 27% identified training as the most critical thing they needed.
This is where practical training has a role. The learning challenge is not simply understanding what AI can do. It is building the judgement to frame the task clearly, choose what information to provide and check what comes back before it is used in real project work. But while that gap stays unsolved, people spend hours on documentation AI could help organise much faster. Risk signals stay buried in email threads until they've become actual issues. The weekly sponsor update is still a scramble through Teams chats and half-remembered decisions.
The admin burden rarely looks dramatic in the moment. It shows up as the second version of the sponsor update, the action log no one has cleaned up, the risk mentioned in passing and the follow-up email that takes longer than the meeting.
The update AI generates is only ever as good as the tracking and workflow details sitting behind it. A status report built on outdated information is still an outdated status report, regardless of how well it's written. Real capability means keeping the tracking current, so what goes into AI is accurate. It also means doing something with what comes out: acting on flagged actions, following up on risks and making the decisions the update has put on the table.
Where AI Actually Saves Time
It shows up in the everyday work first. Pre-meeting preparation can become faster with a clear prompt and current project information. The status update covers what sponsors need to know rather than everything that happened. Workshop outputs get written up before people have left the building rather than three days later when the details have gone cold.
When the admin layer is lighter, project leads have more space to notice when a stakeholder’s tone has shifted, assess a timeline risk properly and schedule the decision everyone knows needs to happen before the next milestone. None of that shows up in a prompt. It comes from being close enough to the work to read it.
The people trusted with the next project are usually the ones who keep stakeholders informed, reduce surprises and limit the chasing. AI will not earn that trust for them. But it can clear enough of the reporting and follow-up work for them to stay closer to the people, risks and decisions that matter.
What This Means for Project Professionals
For project professionals, the opportunity is not to hand the project over to AI. It is to reduce the manual work that keeps pulling attention away from delivery. The relationships, trade-offs and judgement calls still sit with the person leading the project.
AI will not improve project outcomes for people who try it once, get a generic output and return to manual work. AI becomes useful when project leads know how to prompt AI, evaluate what it returns and recognise when human judgement needs to take over.
AIM's AI for Project Management course helps project professionals build practical AI capability across project updates, meeting notes, risk tracking and stakeholder communication, while keeping human judgement at the centre of the work.
For professionals ready to take the next step in their career, now is the time to act. Enrol before June 30 and receive 20% off AI for Project Management.
