I get this question a lot, and it usually reveals a misunderstanding worth clearing up first: "Python" and "AI" aren't two separate things you choose between. Python is a programming language. AI is a field. For some AI careers you need Python; for others you don't need it at all. So the real answer depends on which AI career you're targeting. Let me break it down.

If you want a non-technical AI role: learn AI tools, skip Python

If your goal is AI content strategy, prompt engineering, AI product management, or AI consulting, you don't need Python at all. Learning it would be a detour. Spend your time getting deeply fluent with AI tools — ChatGPT, Claude, image models, automation platforms — and building a portfolio. Python would be wasted effort for these roles.

This is the path most career-changers should take, because these roles are more accessible and your existing domain expertise transfers directly. Don't learn a programming language you'll never use.

If you want to be an ML engineer: learn Python first, then AI

If your goal is genuinely to build AI systems — to be an ML engineer or AI developer — then yes, learn Python first. Python is the dominant language for machine learning. You need solid programming fundamentals before the AI-specific concepts (model architectures, training, evaluation) will make sense.

The sequence matters here. Trying to learn ML concepts before you can program is like trying to write poetry before you know the language. Get comfortable with Python — variables, functions, data structures, working with libraries — then layer the ML concepts on top.

The realistic Python-first sequence

Free resources for this whole path exist: Python via the official tutorial or freeCodeCamp, ML via fast.ai (genuinely world-class and free), Kaggle Learn for hands-on practice. You can do the entire sequence without paying for anything.

The test to decide which path is yours

Ask yourself one question: do you want to build AI systems, or use AI systems to do valuable work? If you want to build them, learn Python first. If you want to use them, skip Python and get fluent with the tools. Most people who ask this question actually want the second thing — they want to do more valuable work with AI — and they'd be wasting months learning Python they don't need.

What I'd tell a friend

When friends ask me this, I push back with a question: what do you actually want to do day to day? Nine times out of ten, when they describe their real goal, it's a role that doesn't require Python. They'd been told "learn to code" as generic advice, when the better advice for their actual goal was "get really good at using these tools and apply them to what you already know."

The one in ten who genuinely want to build models — those should learn Python first, no question.

The honest take

Don't learn Python because you heard you should. Learn it only if your target role actually requires it — which means ML engineering and AI development. For the much larger category of AI-adjacent roles, Python is a distraction. Figure out which career you actually want first, then the Python-or-AI question answers itself. For most career-changers, the answer is: skip Python, master the tools.

Sources