AI in Project Management: Where It Helps, Where It Bluffs
AI tools can draft a status report in seconds. They can also state a wrong date with total confidence. For a project manager, the useful question is not whether to use AI but which tasks to hand it and how to check the result. This guide covers what a recent survey says about adoption, seven tasks where AI helps, its limits, a simple rule for what data to paste in, a time-savings example with the checking time included, and PMTQ, the technology skill set PMI has described.
The short answer
AI is good at first drafts and summaries of text you give it. It is weak at knowing your project: the side agreement, the stakeholder who hates surprises, the date someone promised in a hallway.
So use it to produce drafts, then check them. You stay accountable for what goes out under your name.
The saving is real but smaller than the marketing suggests, because checking takes time. The example in section 5 shows how to count it.
If you are still learning the basics of the role, see what is project management first.
What the survey says about adoption
In September 2025 the Association for Project Management (APM), a UK professional body, published a survey of 1,000 project professionals carried out by Censuswide. It reported that 70 percent said their organization currently uses AI, up from 36 percent in 2023. Another 29 percent said their organization plans to adopt it.
Among respondents whose organizations use AI and who had seen a benefit, about half named task and schedule automation, resource allocation, and risk analysis and forecasting. About 49 percent named reporting and dashboards, and 43 percent named stakeholder communications.
The same survey recorded concerns. Among people using AI, 49 percent cited technical knowledge and training, 44 percent security and data privacy, and 41 percent inaccuracy or untrustworthiness.
Read it with care. It is a self-reported survey from one country, and it measures what organizations use, not how well it works. It does not show that AI improves project results.
Seven tasks: where AI helps
This table is our assessment of where AI fits today. Tools change quickly, so test them on your own work.
| Task | How AI helps | Check before you trust it |
|---|---|---|
| 1. Meeting notes and action items | Turns a transcript or your rough notes into a clean summary | Names, owners and dates against what was said |
| 2. Status report drafts | Writes a first version from your task list and your own notes | Every number, and anything that sounds more positive than your own view |
| 3. Risk brainstorming | Suggests risks you did not think of, by project type | Which ones actually apply to your project |
| 4. Estimate ranges | Offers a first range and the questions to ask | Compare with your own past projects. Do not use as the estimate. |
| 5. Stakeholder emails | Rewrites for tone and length, or drafts a first version | Tone, promises made and anything that commits you |
| 6. Requirements wording | Turns notes into clear requirement or user story drafts | That it kept the meaning the stakeholder gave |
| 7. Lessons learned | Groups comments from a retrospective into themes | That a minority view was not lost in the summary |
Some project tools also use machine learning on schedule and cost history to flag likely slips. That can be useful, but it depends on how good your project data is. Poor data gives poor warnings. Our guide to useful project metrics covers what to track.
Limits, and a rule for what data to paste in
Four limits to keep in mind:
- It can be wrong and sound sure. Dates, names, figures and citations are common places for mistakes.
- It does not know what you did not tell it. Informal agreements and politics are invisible to it.
- It cannot be accountable. If a bad report goes out, you answer for it.
- Data can leave your control. What you paste into a tool may be stored or used in ways your company has not approved.
For the last point, use a simple traffic-light rule. Always follow your employer’s own policy first, because rules differ.
| Level | Examples | Rule |
|---|---|---|
| Green | Generic questions, public information, text you wrote with no names or figures | Usually fine in any approved tool |
| Amber | Internal plans, team names, schedules, anonymized status data | Only in tools your company has approved. Remove names where you can. |
| Red | Customer data, contracts, pricing, personal information, unreleased financials | Do not paste unless your company has cleared that tool for it in writing |
If your employer has no policy, ask before you start. That conversation is worth having.
A worked example: weekly time, with checking included
This example is made up. Say a project manager spends four hours and 15 minutes a week on four writing tasks. The table shows time before and after using AI for drafts, with the time to check and fix the output counted.
| Weekly task (made-up) | Before (min) | With AI draft plus checking (min) | Saved (min) |
|---|---|---|---|
| Status report | 90 | 55 (20 to prepare and generate, 35 to check and fix) | 35 |
| Notes from three meetings | 60 | 45 (3 × 15) | 15 |
| Stakeholder emails | 60 | 45 | 15 |
| Risk register review | 45 | 45 (more ideas, same time to filter) | 0 |
| Total | 255 | 190 | 65 |
That is 65 minutes a week. Over 48 working weeks it comes to 3,120 minutes, or 52 hours.
Now test the assumption that matters most. If checking the status report takes 70 minutes instead of 35, that task takes 90 minutes, the same as before, and the weekly saving falls from 65 minutes to 30.
The lesson is to measure your own checking time for two weeks before you plan around any savings. The figures above are for illustration and say nothing about what you will save.
PMTQ: the technology skill set
If you came here looking for PMTQ, this is where it fits. In a 2019 article, the Project Management Institute (PMI) described technology quotient (TQ) as a person’s ability to adapt, manage and integrate technology. It used PMTQ for TQ combined with project management. The article was written before today’s generative AI tools, so treat it as a framing, not a current guide.
PMI named three characteristics of a high PMTQ:
- Always-on curiosity. Looking for new ways to deliver, while staying skeptical enough not to chase every trend.
- All-inclusive leadership. Getting the best from a team regardless of age, rank, digital skill or location.
- A future-proof talent pool. Building a team that keeps its skills current and helps others do the same.
A quick self-check, in our own words:
- Did I try one new tool or method on a real task in the last three months?
- Do I know which team members are comfortable with new tools, and who needs support?
- Can I explain, in two sentences, what a tool does and where it fails?
- Do I have a way to check the output before it goes out?
Technology skills sit alongside people skills. See soft skills for project managers for the other half.
A 30-day starter plan
| Week | Do this |
|---|---|
| 1 | Find out what your employer allows. Pick one green-level task, such as rewriting an email you already wrote, and try it. |
| 2 | Time one weekly task with and without a draft, including checking. Write down both numbers. |
| 3 | Keep a log of errors the tool made. Note the kind: dates, names, tone, invented detail. |
| 4 | Decide which tasks to keep using it for, add a checklist for each, and tell your team what you do and do not use it for. |
Common mistakes
- Sending a draft unread. Always read every line that goes out under your name.
- Pasting confidential data. Check the policy before you paste anything in the amber or red levels.
- Using it as the estimator. An invented number looks the same as a real one.
- Counting the saving without the checking. The example shows how fast it can disappear.
- Hiding it from the team. If people find out later, it can cost you trust.
- Buying tools before fixing the data. Forecasts from messy project data are not reliable.
Learn more
If you want structured learning, online providers such as Coursera and Udemy list project management courses, and some include modules on tools and AI. We may earn a commission if you sign up, at no cost to you. We are not affiliated with any university. Check the syllabus and the date it was last updated, because this area changes fast. To go back to the fundamentals, read project manager jobs.
Frequently asked questions
Will AI replace project managers?
We cannot know. Today it handles drafting and summarizing well but not accountability, negotiation or judgment about people. Treat confident predictions in either direction with caution.
What can AI do in project management?
Draft status reports, summarize meetings, suggest risks, rewrite emails, turn notes into requirements and group retrospective comments. See the seven-task table above.
Is it safe to paste project information into an AI tool?
It depends on the tool and your employer’s policy. Use the green, amber and red rule above, and ask before you paste anything sensitive.
What is PMTQ?
PMI’s term, from 2019, for technology quotient applied to project management: the ability to adapt, manage and integrate technology on projects.
Do I need a certification in AI?
Not to start. Most of the value comes from practicing on real tasks and checking the results. If you want a course, compare the syllabus and the date it was updated.
How do I get started?
Pick one low-risk task, time it with and without a draft, count the checking, and keep a log of errors. The 30-day plan above gives a path.
Browse certificate, course and degree programs side by side.