01Plan
Get from blank page to a defensible starting point faster.
- Project plans
- Work breakdown
- Requirements
- Research
AI for PMs
AI won't replace project managers. But it is changing how we work.
The goal is not to use AI for everything. It's to know where it genuinely helps, where your judgment still decides, how to check what it produces, and how to fit it into the project workflows you already run.
Four things to get right
Practical, not futuristic.
Where AI helps
Drafting, summarising, restructuring, pattern-spotting and repeatable admin — the work that eats your week but rarely needs your judgment.
Where human judgment matters
Trade-offs, escalation, politics, commitments and anything a stakeholder will hold you to. AI can prepare the thinking; it cannot own the decision.
How to verify output
Check it against the source artifact, test the numbers, look for confident-sounding gaps, and never send anything you couldn't defend in a steering meeting.
How to integrate it
Fit AI into the workflow you already run — your status cycle, your RAID log, your reporting cadence — instead of bolting on a separate tool nobody maintains.
Where it applies
Five places AI earns its seat on a project.
02Deliver
Keep the delivery machine running without drowning in admin.
- Meeting intelligence
- Action tracking
- RAID management
- Workflow support
03Communicate
Say the right thing, to the right audience, at the right altitude.
- Status reporting
- Stakeholder communication
- Executive updates
- Presentation support
04Think
Use AI as a thinking partner — then apply your own judgment.
- Risk analysis
- Scenario planning
- Decision support
- Pattern identification
05Automate
Turn the things you repeat every week into something that runs itself.
- Repeatable project workflows
- AI assistants
- Agents
- Scheduled project intelligence
Responsible AI
Humans in the loop. Always.
Every AI-assisted artifact on a project still needs a named human owner. You review before it goes out, you check the facts against the source, and you stay accountable for what it says.
That means being careful with client and employee data, understanding your organisation's policy before you paste anything into a tool, and being honest with stakeholders about how work was produced.
Done properly, this is what makes AI adoption stick on a project — trust, not novelty.
Where to start
AI-Ready PM Self-Assessment
Ten questions to gauge where you actually are — with personalised next steps.
OpenAI Delivery Maturity Assessment
A deeper look at how AI is showing up across your delivery practice.
OpenAI in Project Management class
A five-week live class applying AI to your real project work.
OpenRolling AI out across a team or PMO? See organisational training.