Things You Should Do With AI as a Product Manager

Your job as a Product Manager is to shape the product and create value. Use AI for the work that slows you down. Focus on what actually matters.

As a Product Manager, your value is direction and trade-offs. Your job is to shape the direction of the product. You decide what gets built and why. You make sure your team creates value. Everything else is support work. Use AI for that. Free your time. Spend it where it counts.

The hard rule

AI can prepare the thinking and challenge your ideas. It cannot own decisions.
Use AI for work that is repetitive, language-heavy, manually time-consuming, low-risk if it is 80 percent right. AI helps where you process input, not where you choose.

6 Things you should do with AI

1. Turn chaos into a decision you can actually make

Use AI to turn messy notes, opinions, and data into a clear decision setup. Let it structure the decision, list real options, surface trade-offs, and point out what information is missing so you can actually decide instead of circling.

2. Backup your communication with a story

Most PM communication fails because it's a list, not a story.

Use AI to force your thinking into a clear story: why now, what problem matters, what happens if you do nothing, and what decision you want. You decide on a message, AI removes the rambling.

3. Find contradictions in your own thinking

Many PMs carry 12 mental models at once and don't notice conflicts.

Use AI as a mirror. Feed it your strategy, roadmap, and key arguments and ask it to point out contradictions, hidden assumptions, and places where your actions don't match your stated goals. This is about exposing blind spots before others do.

4. Turn feedback into something you can act on

As a PM you get endless feedback. And if you're unlucky, it is either vague or emotional.

Use AI to process raw feedback into something usable. Let it reframe emotional or vague feedback into concrete problems, group similar signals, and highlight patterns so you stop reacting to noise and start seeing what actually repeats.

5. Cut down scope

Have you ever gone through this? You ask engineering for "small change" and it explodes. Use AI to prepare the scope with you. Let it analyze your idea and split the work into must-haves from nice-to-haves. Build a good-better-best ladder.

6. Communicate, communicate, communicate

If you avoid communication, it turns into conflict. We know writing emails sucks. We know you write multiple a day already. You know what's easier? Just ranting about your struggle and let someone handle it for you. Use AI to get hard communication out of your head and into the world.

Literally tell it how you feel using voice input, explain what you need, and ask it to draft the email or message clearly and directly. Review it, send it. You don't even have to leave ChatGPT to do it.

Theresa Hennighausen

About the Author

Theresa Hennighausen

Theresa works on Product Copilot, an AI tool for software product teams. She has seen the product grow from early user interviews to being used by successful product teams. At Prio 0, she works closely with Product Copilot users, gathers feedback, and translates real user needs into practical product decisions. Her focus is on UX, product strategy, communication and designing AI prompts that fit everyday product work.