AI isn’t here to put software engineers out of a job. It’s here to give them a lot more work. The arrival of the Self-Writing Internet isn’t a farewell party for coders; it’s an invitation to a much bigger one, with more people, more ideas, and more opportunities. The idea of machines doing all the heavy lifting might sound like a dystopian fantasy to some, but the reality is much messier, much more human, and far from a story of obsolescence.
The fear that artificial intelligence will displace entire industries is as old as automation itself. Each new wave of technology has had its sceptics, but history has been consistent in proving them wrong. When spreadsheets arrived, accountants weren’t wiped out—they became more efficient, handling more complex financial strategies instead of wasting time on calculations. When digital photography took off, professional photographers didn’t vanish—they adapted, found new ways to work, and reached wider audiences. Software engineering is heading down a similar path.
The idea of machines generating their own code isn’t science fiction anymore. AI-powered development tools can already produce entire applications from a simple prompt. The difference now is scale. The Self-Writing Internet proposes something radical: a future where anyone can become a software creator, not by learning to code, but by learning to communicate with machines in a way they understand.
That’s where prompt engineers come in. If you’ve ever found yourself tweaking the wording of a search query to get better results, you’ve already done a basic version of what a prompt engineer does. The difference is that instead of searching for information, prompt engineers guide AI systems to generate applications, automate processes, and solve problems. It’s not about replacing human thought, but about refining how humans interact with AI to get meaningful results.
This shift doesn’t eliminate traditional software development. If anything, it expands it. An internet that generates millions of applications on demand doesn’t run itself—those applications need oversight, optimisation, maintenance, and security. They also need professionals who understand how to push them further. Just because an AI can generate a website, for instance, doesn’t mean that website is scalable, secure, or free of bugs. Human developers aren’t being made redundant; they’re being given more to manage.
The demand for tech skills won’t shrink—it will diversify. Instead of a singular path leading to software engineering through years of formal education, new roles will emerge for those who specialise in directing AI, troubleshooting its outputs, and ensuring that what it produces is useful, ethical, and reliable. Writing code will become just one part of a much larger ecosystem, where collaboration between humans and AI is the norm.
The impact of this shift goes beyond the workforce. It changes who gets to participate in the creation of technology. The barrier to entry for building software has traditionally been high, limited to those who invest years into learning programming languages. With AI-driven development, someone with no technical background can launch an app, automate a workflow, or build a business using only their ideas and the right prompts. That kind of accessibility doesn’t just expand the pool of creators—it increases demand for experts who can refine, scale, and secure those creations.
Sceptics might argue that automating development could lower the quality of software or flood the market with half-baked applications. It’s a valid concern, but not a new one. Every major technological shift has come with an explosion of content—some great, some terrible. The web itself was once seen as a chaotic free-for-all, yet it’s now an essential part of daily life, filled with both polished platforms and amateur experiments. AI-generated software will follow the same trajectory: an initial wave of rapid experimentation, followed by a need for professionals to make sense of it all.
For software engineers, this is an evolution rather than an extinction event. Instead of spending hours writing boilerplate code, they’ll focus on refining AI-generated outputs, solving complex problems, and developing new capabilities that AI alone can’t create. There will be more demand for those who understand the intricacies of system architecture, security, and optimisation—fields that automation can assist but not master. The companies that embrace this shift early will be the ones that thrive, hiring engineers who know how to harness AI rather than compete with it.
AI has never been about replacing people; it’s been about changing the nature of work. The Self-Writing Internet will introduce a new wave of entrepreneurs, creatives, and problem-solvers, many of whom would never have considered entering the tech industry before. The future of software development isn’t about machines taking over—it’s about more people than ever being able to build, innovate, and create. Engineers won’t be coding less; they’ll be coding differently, in a world where their skills are more valuable than ever.
