We’re living through an AI gold rush. Every week, a new model, startup, or AI-powered feature launches. Everyone’s claiming their product is “AI-driven.” But here’s the uncomfortable truth: most product managers don’t know what to do with AI.
Some panic: “Will AI replace me?”
Some overhype: “AI will solve everything!”
The truth is somewhere in the middle: AI changes the job, but doesn’t kill it.
The AI Opportunity (If You Play It Right)
AI isn’t magic — but it is leverage. For product managers, this means:
Better signals: AI can crunch customer data, behaviour logs, and feedback at scale to reveal trends humans would miss.
Faster execution: From automating backlog grooming to generating test cases, AI reduces grunt work so you can focus on strategy.
Personalisation at scale: Imagine 1M users, each with a slightly different version of your product tuned to their needs. That’s the real promise.
I once spent two weeks with my team manually tagging thousands of customer support tickets to identify pain points. Today, I could do it in an hour with an off-the-shelf NLP model. The real difference? Back then, we made decisions late. Today, you can act in near real time.
The AI Challenge (What Nobody Wants to Say Out Loud)
Here’s the part that makes PMs sweat:
AI ≠ product-market fit. You can bolt AI onto a broken product, but it won’t save you.
Black box problem: You’ll be responsible for features you can’t fully explain. Customers don’t like “trust me, the algorithm knows.” Neither do regulators.
Bias & ethics: If your AI feature discriminates, misclassifies, or leaks data, guess who’s in the spotlight? Not the data scientist. The product manager.
Skill gap: Most PMs know how to write user stories. Few know how to question a training dataset.
The 7 + 7 Framework for AI Product Managers
Forget hype. Here’s the skill matrix you actually need:
Soft Skills
Strategic Thinking – Don’t chase shiny AI features. Solve real problems.
Communication – Translate AI jargon into business and user value.
Problem-Solving – Frame the right problem before unleashing models.
Ethical Awareness – Ask “Should we build this?” as often as “Can we?”
Data Fluency – You don’t need to code, but you must interrogate data.
User-Centred Design – AI should augment, not confuse the user.
Project Leadership – Keep cross-functional AI teams moving.
Technical Literacy
Programming (Python basics go a long way).
Machine Learning Concepts (supervised, unsupervised, reinforcement).
Data Modelling (where does the data live, and is it clean?).
Natural Language Processing (ChatGPT isn’t magic, it’s NLP at scale).
Computer Vision (think retail, logistics, healthcare).
Cloud & Deployment (AWS, GCP, Azure – where models live).
Data Visualisation (communicate what the AI found, not just raw output).
Opportunities vs. Risks: The Balance Sheet
Opportunities
Automate repetitive tasks
Discover hidden insights
Personalise at scale
Create defensible product moats
Risks
Regulatory landmines (EU AI Act, GDPR)
Overhype leading to customer disappointment
Ethical backfires (bias, privacy)
Dependency on external models you don’t control
Free Resource for Readers
📥 Download the AI-PM Skills Matrix (Free PDF)
A one-page cheat sheet with the 7+7 framework, ready to print or drop in Notion.
AI won’t replace product managers. But product managers who ignore AI will be replaced by those who embrace it.
Your job isn’t to worship AI or fear it; it’s to make better product decisions with it. That means filtering the hype, asking uncomfortable questions, and building products that are useful, ethical, and resilient.
Because at the end of the day, your customers don’t care if it’s AI. They care if it works.
👉 Upgrade to Cooking Agile Premium (€7/month or €70/year) to get my full AI-PM Toolkit: templates for AI use case discovery, stakeholder alignment frameworks, and ethical risk checklists.