It’s the thing we all keep hearing about AI… And what do we keep hearing?

“All PMs need to become AI PMs.”

For my money, it’s a true statement. But with some caveats.

Let me explain.

The 3 types of AI PM

What do we mean by ‘AI PM’?

Actually, there are three types of AI PMs:

• AI-powered PM: This is you, supercharged by AI tools: Writing PRDs in 20 minutes instead of 2 hours and using ChatGPT for competitor research that would take days manually.
• AI PM: These are the specialists building core AI products at companies like OpenAI and Anthropic. They are the ones who ensure ChatGPT doesn’t go rogue. 
• AI feature PM: This is a fast-growing category right now and it’s about adding AI magic to existing products. (Notion’s AI writing assistance, Miro’s smart canvas).

Not every PM will become a specialist in building AI products or features. But every PM needs to become an AI-powered PM (Type One).

And that’s the subject of today’s piece. But how?

The 3 rules to using AI right

The number one thing to realize with AI is that it can’t do the whole job for you. There are three key rules you must remember when using any AI tool:

1. Prompt skill is everything

Each tool has its own nuances for prompting. But generally, there’s a huge skill curve in prompting.

Just like everyone assumes they’re an above-average driver, everyone tends to assume they’re an above-average prompter.

In actuality, you probably have a long way to go as a prompter if you haven’t been practicing.

If you’re using ChatGPT o1 for instance, the optimal prompt structure looks like this from Dan Mac and retweeted by OpenAI’s President Greg Brockman:

2. Leave the middle 60% to AI

You need to do the first 20% of the work in any PM text because the truth is: Claude doesn’t have access to your context as a PM.

So you need to brain-dump the relevant information into it, as the picture above alludes to.

You also need to do the last 20% of the work. That’s where you remove the traces of AI in the text. And you give it your additional unique human context.

That’s why the company has hired you after all! And not just created an AI agent.

3. Revise till you get what you want

It’s rarely the case that the first output you get from an AI is exactly what you want. You generally want to iterate with it.

Provide very clear and specific feedback that it can easily incorporate. It’s usually in drafts 4-5 of a written output that things can be used.

Now let’s get into the specifics…

Top 5 AI PM use cases

There are a million use cases to use AI as an AI-powered PM. Let me share the 5 most game-changing ways I’ve seen PMs transform their work with AI:

1. Leave the heavy lifting to AI when creating PRD

Picture this: It’s 4 PM, and your VP wants a PRD for that new feature by tomorrow morning. Pre-AI you: Panic ordering pizza and pulling an all-nighter vs Post-AI you: Firing up ChatPRD and being done before dinner gets cold.

I recently watched a PM turn a vague idea into a structured PRD in 30 minutes. Their initial prompt structure was simple but effective: 

“Given [feature context], generate a PRD with Problem Statement, User Personas, Success Metrics, and Requirements. Focus on [specific use case].”

With some well-written context and iterations, the draft was taken to 80% in 10 minutes. Then, they spent 20 minutes putting human touches on things they could only do.

The best part? The AI handles the heavy lifting, while you focus on adding the insights only you can provide.

2. Finalize your strategy docs

Product strategy is one of the most important jobs for PMs. But, a lot of times, your strategy document is more about the research and work that you do before it.

When it comes time to actually writing a hard-hitting doc, the highest returns aren’t always in writing it beautifully. That’s where a tool like Claude comes in. If you give it all the right thinking and core principles, it can come back with a well-defended articulation of them.

The prompt structure that works best here is to overwhelm with context, where you have 5-6 points of key context so that Claude doesn’t misunderstand your situation. Something like this:

There aren’t many software product strategies in its training data, so feeding it a great one from your company also helps.

3. Accelerate your competitor research

We’ve all been there: a new competitor comes out of nowhere and shakes up the market. You know your bosses are going to be looking for your take on it.

Perplexity can vastly speed up your time to insights. It’s like Google search enhanced.

Here’s how you can do a competitive research with it:

1. Feed it that new competitor.
2. Ask for feature comparisons, pricing strategies, and user sentiment.
3. Let it dig through recent updates you might have missed.

Here’s an example of a report it created for a hypothetical PM at Lemlist about Instantly.

AI might not be able to go through everything you need to as a PM, but it makes a 3-hour exercise – writing up some thoughts to your boss about a new competitor – more like 30 minutes. 

AI can do most of the fact-finding and point you to links with more info. You can add the thinking on top. 

4. Prepare for the meetings and impress executives without breaking a sweat

We’ve all been there – scrambling before an exec review, trying to remember every action item from last time.

Here’s what changed the game for me: Feed ChatGPT your last meeting notes (if your company doesn’t pay for transcription use a note-taking tool), and watch it create:

• Progress updates for each action item.
• Data-backed responses to previous concerns.
• A structured agenda for the next meeting.

My favorite part? Executives think I spend hours preparing. Reality: 15 minutes with AI while grabbing coffee.

5. Scale prototyping and design iteration

“But can it help with design?” Oh, just watch.

Last week, I saw a PM go from an idea to a clickable prototype in under an hour. Their secret? Claude Artifacts for wireframes.

Don’t believe me?

Here’s a full clickable prototype Claude created in 15 seconds with a simple prompt: “Please build a clickable prototype of a calendar app.”

The game-changer isn’t just speed – it’s iteration. When stakeholders want changes, you’re not redoing hours of work. You’re tweaking a prompt and generating new versions in minutes.

Real talk: The first versions won’t win design awards. But for early feedback and concept validation? Pure gold.

🚫 Let’s talk about mistakes

I’ve made them all, so you don’t have to.  Here are the face-palm moments every AI PM needs to avoid:

1. Treating AI as magic

A PM I know once let AI write their entire product strategy without any human input. Spoiler alert: The review meeting was… interesting.

The strategy looked perfect on paper, but missed crucial context about:

• Their company’s unique culture.
• Past failed experiments.
• Office politics (yes, that matters).

Lesson learned? AI is your co-pilot, not your autopilot.

2. The “Send it right away” trap

Picture this: Late night, big presentation tomorrow. AI generates your slides. They look good. You hit send without checking.

Next morning: You’re explaining to your CEO why your market size numbers are from 2019 and your competitor analysis includes a company that got acquired last month.

True story. Happened to a PM who messaged me.

The fix? My 5-minute validation ritual:

• Fact-check any specific numbers.
• Verify competitor info is current.
• Cross-reference market trends.
• Sanity-check recommendations.

3. The tool time capsule

A quick story: A PM spent months using an outdated version of ChatGPT for user story generation. Meanwhile, Claude was doing the same job in half the time with better results.

The AI landscape changes weekly. What was best-in-class last quarter might be outdated today.

My reality check routine:

• Test new AI tools monthly.
• Reevaluate your tech stack quarterly.
• Always ask other PMs what’s working for them.

4. The silent PM syndrome

Here’s a painful one: A PM found an amazing AI workflow for user research analysis. Did they share it with their team? Nope. Result? Four PMs, same company, all solving the same problem differently.

AI knowledge compounds when shared. Start a simple Slack channel. Share your winning prompts, tool discoveries and epic fails (yes, especially these).

5. The all-or-nothing trap

The most dangerous type of PM? The AI maximalist. You know the one: “If AI can’t do everything, why bother using it at all?”

Wrong. So wrong.

Here’s the better approach: start small and scale up.

• Week 1: Use AI for meeting notes.
• Week 2: Add PRD outlines.
• Week 3: Experiment with competitive research.

Build confidence step by step.

Remember, not everything needs AI, and that’s perfectly fine. Focus on what adds value, not on forcing AI into every task.

Remember: Every PM I know who’s crushing it with AI started by failing fast and learning faster.