Everyone is using AI now. Or at least, everyone says they are.
But most project managers are using it the same way they'd use Google — typing vague questions and getting vague answers. "Help me write a project update." And Claude or ChatGPT gives them something generic that still needs to be completely rewritten.
That's not AI saving you time. That's AI adding a step.
The problem is the prompt, not the tool.
AI is only as useful as the instruction you give it. A vague prompt gets a vague output. But a specific, structured prompt — one that includes your project context, your audience, and the exact format you need — gets you something you can use in under 60 seconds.
Most project managers don't know how to write those prompts. Not because they're not smart, but because nobody taught them. Prompt engineering sounds technical. It's not. It's just knowing what to ask for and how to ask for it.
What AI can actually do for your projects:
When prompted correctly, AI can write your entire weekly stakeholder update in 45 seconds. It can summarise a 2-hour meeting transcript into 5 bullet points and a decision log. It can identify risks in a project brief you paste in. It can draft a scope change request email that actually gets approved. It can create a vendor comparison table from a list of criteria you provide.
None of this requires technical skills. It requires the right prompts.
The AI Prompt Library — PM Edition
I spent weeks testing and refining 20 Claude and ChatGPT prompts specifically for project management tasks. Each prompt is copy-paste ready — you open the tool in your browser, click copy, paste into AI, fill in your project details, and get a usable output.
The 20 prompts cover: status reports, risk identification, stakeholder emails, meeting summaries, scope change requests, vendor briefings, project kickoff agendas, and more.
People using these prompts consistently report saving 8–12 hours per week on PM admin tasks alone.
— Arnie Rose Felicilda, Supply Chain & Project Management Educator
