I don’t think AI is replacing programmers anytime soon — but it’s getting closer than most people realize. Every few months a new tool comes out that can write code, analyze documentation, and even scaffold entire projects. AI tools can now pass technical coding interviews, generate working components from screenshots, and write code that’s often indistinguishable from junior developer output.
AI still can’t take a prompt and deliver a complete, production-ready project. Even when it picks the right language or helps write boilerplate, the intermediate steps — testing, debugging, performance tuning, making architectural decisions — that’s where it falls short. And those are the parts that matter most.
What GitHub’s CEO Has to Say
GitHub’s CEO Thomas Dohmke has been pretty clear on this — AI won’t replace developers, it’ll make them faster. And GitHub put their money where their mouth is with Copilot, an AI pair programmer trained on massive amounts of code. The goal isn’t to replace you — it’s to handle the repetitive parts so you can focus on the harder problems.
Dohmke also points out something important: there aren’t enough developers to meet demand. The shortage of computer science talent is real, and as AI tools become more common, the demand for developers who understand AI, APIs, and model training is only growing.
The Numbers Back It Up
That gap is telling. The more specialized and creative the work, the harder it is to automate. WordPress theme development sits squarely on the creative side — you’re making design decisions, solving UX problems, and writing code that integrates with a massive ecosystem of plugins and hosting environments. That’s why I’m confident that developing WordPress Themes is a viable and profitable way to earn a living for the foreseeable future.
How I Actually Use AI in My Workflow
I use AI tools regularly — Copilot for code suggestions, ChatGPT for brainstorming and debugging, and various AI-assisted analysis tools. They’re genuinely useful. But every suggestion still needs a human to evaluate whether it’s correct, whether it fits the architecture, and whether it introduces security issues. AI is a fast assistant that’s wrong often enough that you can’t trust it blindly.
If you want to learn how to use ChatGPT effectively, I put together a ChatGPT Tutorial for Beginners that covers the basics.
So Will AI Replace Programmers?
Not anytime soon.
The developers who learn to work with AI will be more productive than those who ignore it, and way more productive than the AI working alone.
Check out the full discussion in the video:
For more on how AI fits into WordPress specifically, check out my article on using AI with WordPress.
AI & Programming FAQ
What programming skills are hardest for AI to replace?
Architecture decisions, security auditing, and debugging complex systems across multiple services. AI can generate code, but it can’t reason about why a system fails under load or how a change affects 15 other components. The more context-dependent the work, the more it needs a human.
Should new developers still learn to code if AI can write code?
Absolutely. AI generates code — it doesn’t understand it. Someone still needs to know whether the output is correct, secure, and maintainable. Learning to code gives you the foundation to evaluate AI output critically instead of shipping whatever it suggests.
Will AI make senior developers more valuable or less?
More valuable. Senior developers spend most of their time on design, review, and debugging — exactly the tasks AI handles poorly. AI automates the parts seniors already delegate to juniors, which means seniors who can direct AI effectively become a force multiplier for their entire team.