As AI systems expand their already impressive capacities, there is an increasingly common belief that the field of computer science (CS) will soon be a thing of the past. This is being communicated to today’s prospective students in the form of well-meaning advice, but much of it amounts to little more than hearsay from individuals who, despite their intelligence, speak outside of their expertise.

High-profile figures like Nobel Prize-winning economist Christopher Pissarides have made this argument, and as a result, it has taken root on a much more mundane level – I have even personally heard high school careers advisers dismiss the idea of studying CS outright, despite having no knowledge of the field itself.

These claims typically share two common flaws. First among them is that the advice comes from people who are not computer scientists. Secondly, there is a widespread misunderstanding of what computer science actually involves.

AI and the myth of code replacement

It is not wrong to say that AI can write computer code from prompts, just as it can generate poems, recipes and cover letters. It can boost productivity and speed up workflow, but none of this eliminates the value of human input.

Writing code is not synonymous with CS. One can learn to write code without ever attending a single university class, but a CS degree goes far beyond this one skill. It involves, among many other things, engineering complex systems, designing infrastructure and future programming languages, ensuring cybersecurity and verifying systems for correctness.

AI cannot reliably do these tasks, nor will it be able to in the foreseeable future. Human input remains essential, but pessimistic misinformation risks steering tens of thousands of talented students away from important, meaningful careers in this vital field.

What AI can and can’t do

AI excels at making predictions. Generative AI enhances this by adding a user-friendly presentation layer to internet content – it rewrites, summarises and formats information into something that resembles a human’s work.

However, current AI does not genuinely “think”. Instead, it relies on logical shortcuts, known as heuristics, that sacrifice precision for speed. This means that, despite speaking like a person, it cannot reason, feel, care, or desire anything. It does not work in the same way as a human mind.

Not long ago it seemed that ‘prompt engineering’ would replace CS. Today, however, there are virtually no job postings for prompt engineers, while companies like LinkedIn report that the responsibilities of CS professionals have actually expanded.

Where AI falls short

What AI provides is more powerful tools for CS professionals to do their jobs. This means they can now take concepts further – from ideation to market deployment – while requiring fewer support roles and more technical leadership.

There are, however, many areas where specialised human input is still essential, whether for trust, oversight or the need for human creativity. Examples abound, but there are 10 areas that stand out in particular:

Adapting a hedge fund algorithm to new economic conditions. This requires algorithmic design and deep understanding of markets, not just reams of code.
Diagnosing intermittent cloud service outages from providers like Google or Microsoft. AI can troubleshoot on a small scale, but it cannot contextualise large-scale, high-stakes troubleshooting.
Rewriting code for quantum computers. AI cannot do this without extensive examples of successful implementations (which do not currently exist).
Designing and securing a new cloud operating system. This involves high-level system architecture and rigorous testing that AI cannot perform.
Creating energy-efficient AI systems. AI cannot spontaneously invent lower power GPU code, or reinvent its own architecture.
Building secure, hacker-proof, real-time control software for nuclear power plants. This requires embedded systems expertise to be mixed with the translation of code and system design.
Verifying that a surgical robot’s software works under unpredictable conditions. Safety-critical validation exceeds AI’s current scope.
Designing systems to authenticate email sources and ensure integrity. This is a cryptographic and multi-disciplinary challenge.
Auditing and improving AI-driven cancer prediction tools. This requires human oversight and continuous system validation.
Building the next generation of safe and controllable AI. Evolving towards safer AI cannot be done by AI itself – this is a human responsibility.

Why Computer Science is still indispensable

One thing is certain: AI will reshape how engineering and Computer Science is done. But what we are faced with is a shift in working methods, not a wholesale destruction of the field.

Whenever we face an entirely new problem or complexity, AI alone will not suffice for one simple reason: it depends entirely on past data. Maintaining AI, building new platforms, and developing fields like trustworthy AI and AI governance therefore all require CS.

The only scenario in which we might not need CS is if we reach a point where we no longer expect any new languages, systems, tools, or future challenges. This is vanishingly unlikely.

Some argue that AI may eventually perform all of these tasks. It’s not impossible, but even if AI became this advanced, it would place almost all professions at equal risk. One of the few exceptions would be those who build, control, and advance AI.

There is a historical precedent to this: during the Industrial Revolution, factory workers were displaced at a 50 to 1 ratio as a result of rapid advances in machinery and technology. In that case, the workforce actually grew with a new economy, but most of the new workers were those who could operate or fix machines, develop new machines, or design new factories and processes around machinery.

During this period of massive upheaval, technical skills were actually the most in-demand, not the least. Today, the parallel holds true: technical expertise, especially in CS, is more valuable than it ever has been.

Let’s not confuse the next generation with the opposite message.