AI for Project Management

TL;DR

• AI can draft status updates, meeting notes, sprint plans, and capacity templates.
• It can't read the room, handle politics, or make judgment calls on prioritization.
• Use AI for the paperwork. You own the relationships and decisions.

Project management is 30% process and 70% people. AI handles a chunk of the 30%. The 70% is still you.

Status Reports and Updates

Good use cases:

• "Turn these bullet points into a 3-paragraph status update for leadership"
• "Summarize this week's progress. Tone: concise, no jargon"
• "Draft a 'what's blocked' section from these notes"

What to add:

• Stakeholder nuance ("VP doesn't want to hear about X")
• Political framing ("We're not calling this a delay, we're calling it a reprioritization")
• Things you're not saying (sensitive info, confidence levels)

AI gives you a draft. You make it real.

Sprint Planning and Capacity

Good use cases:

• "Draft sprint goals from these ticket titles"
• "Suggest how to break this epic into stories"
• "Create a capacity planning template for 6 engineers"

What AI can't do:

• Know who's on vacation, who's overloaded, who needs growth opportunities
• Balance "urgent" vs. "important" in your org
• Predict how long things actually take (historical data beats AI)

Use AI for structure. You fill in the humans.

Meeting Notes and Action Items

Good use cases:

• "Summarize this transcript. Extract action items and owners"
• "Turn this messy notes doc into structured meeting notes"
• "Create a follow-up email from this meeting"

Cautions:

• AI may miss nuance. "We'll try" vs. "We will" — different commitments.
• Sensitive conversations — don't paste anything confidential.
• Verify action items. AI can hallucinate a "we agreed to X" that didn't happen.

Risk and Blocker Documentation

Good use cases:

• "Draft a risk register from these bullet points"
• "Write a 'blockers and mitigations' section for this project"
• "Create a RAID log (Risks, Assumptions, Issues, Dependencies)"

AI structures well. You know what's actually a risk vs. paperwork.

Specs and Requirements

Good use cases:

• "Turn this conversation into a PRD outline"
• "Draft user stories from this feature description"
• "Create acceptance criteria for this ticket"

Again: draft. You add business context, edge cases, and "what did we not think of?"

Do This Next

1. Use AI to draft one status update this week. Compare to what you'd write. Add the human layer.
2. Create a "sprint planning" prompt template — inputs: team size, epic summary, constraints. Output: draft sprint goals. Customize and reuse.