Home/Knowledge Base/Knowledge Worker Playbooks/Project Manager

PATHKnowledge Worker Playbooks · Learning path

AI for Project Managers

Projects run on information — status, risks, decisions, lessons. AI can finally read all of it. Whether that helps or embarrasses you depends on the state of your project site.

The role today

Project managers spend a startling share of the week producing information about the project rather than moving the project: status reports, risk registers, meeting notes, stakeholder updates. It’s necessary work, and it’s exactly the kind of structured, repetitive writing AI does well — provided the underlying record is trustworthy.

How AI changes this role

AI helps you with

  • Status report first drafts from the project record
  • Meeting summaries and action tracking
  • Risk register drafting and updates
  • Lessons-learned synthesis at close-out

You provide

  • Stakeholder trust and difficult conversations
  • Judgement calls under uncertainty
  • Negotiating scope, budget and priorities
  • Reading the room when the data says fine but the team doesn’t

This is The Knowledge Worker AI Framework™ applied to this role — AI takes the production, you keep the bookends.

Liza’s take

Twenty years of SharePoint consulting is twenty years of working inside other people’s projects. The pattern never changes: the projects that run well have a single, current, structured home for their information. The ones that don’t have three versions of the plan and a PM doing archaeology every Friday afternoon.

Where AI genuinely helps

Friday reporting in minutes

If the plan, risks and actions live current in one project site, AI can draft the status report from them. The PM who maintains a clean record gets Fridays back.

Meetings that produce a record

Summaries with owners and dates from every recorded session — verified by you, filed in the site, findable in six months.

Lessons learned that actually get read

AI can synthesise a project’s history into honest close-out themes — but only if the history was captured as you went.

Early risk signals

Ask questions across your own project record: what’s slipped twice, which actions keep rolling over. The answers were always there; now they’re a query away.

A day with AI

Monday: draft status report generated from the project site; you correct the tone on one slipped milestone before it goes anywhere. Midweek: a stakeholder asks what was decided about the vendor in March — a question you answer with a filter, not a search party, because decisions are logged with metadata. Friday: the risk workshop’s recording becomes a draft register update; you strike two items the AI over-weighted because it can’t feel the team’s confidence. That last part is your job.

The risks

A confident report built on stale data is worse than no report — if the project site holds superseded plans, AI will present them beautifully. Watch permissions too: project sites accumulate guests and old shares, and anything a person can access, AI can surface to them. And never let generated minutes stand as the record of a contested decision without your verification.

Where the role is heading

Coordination is compressing; judgement is not. I expect PMs to spend less time producing information and more time on the human system around the project — and the PMs who also own the project’s information architecture will be the ones organisations fight to keep. Watch for titles like Delivery Lead and Project Intelligence Manager.

Your learning path

Start with the cards below, in order — each one is a five-minute read built for a task this role repeats. The full K-Series (ten foundations, ten workflow cards) lives on the Knowledge Worker Playbooks hub.

Can AI write my status reports?

It can draft them well — from a current, structured project record. If your plan and risks live in scattered files and inboxes, fix that first; the report quality is downstream of the record quality.

Is it safe to use AI on project documents?

Use the tools your organisation has approved, keep commercial and personal data out of anything unsanctioned, and check who has access to your project site — AI surfaces whatever permissions allow.

What’s the highest-value AI habit for a PM?

Verified meeting capture. A project whose meetings reliably produce owned, dated, filed actions compounds week after week — and it’s the raw material every other AI assist depends on.