Some of Julie York’s high school computer science students are worried about what generative artificial intelligence will mean for future careers in the tech industry. If generative AI can code, then what is left for them to do? Will those jobs they are working toward still be available by the time they graduate? Is it still worth it to learn to code?

They are “worried about not being necessary anymore,” said York, who teaches at South Portland High School in South Portland, Maine. “The biggest fear is, if the computer can do this, then what can I do?”

The anxieties are fueled by the current landscape of the industry: Many technology companies are laying off employees, with some linking the layoffs to the rise of AI. CEOs are embracing AI tools, making public statements that people don’t need to learn to code anymore and that AI tools can replace lower or mid-level software engineers.

However, many computer science education experts disagree with the idea that AI will make learning to code obsolete.

Technology CEOs “have an economic interest in making that argument,” said Philip Colligan, the chief executive officer of the Raspberry Pi Foundation, a U.K.-based global nonprofit focused on computer science education. “But I do think that argument is not only wrong, but it’s also dangerous.”

While computer science education experts acknowledged the uncertainty of the job market right now, they argued it’s still valuable to learn to code along with foundational computer science principles, because those are the skills that will help them better navigate an AI-powered world.

Why teaching and learning coding is still important, even if AI can spit out code

The Raspberry Pi Foundation published a position paper in June outlining five arguments why kids still need to learn to code in the age of AI. In an interview with Education Week, Colligan described them briefly:

1. We need skilled human programmers who can guide, control, and critically evaluate AI outputs.
2. Learning to code is an essential part of learning to program. “It is through the hard work of learning to code that [students] develop computational thinking skills,” Colligan said.
3. Learning to code will open up more opportunities in the age of AI. It’s likely that as AI seeps into other industries, it will lead to more demand for computer science and coding skills, Colligan said.
4. Coding is a literacy that helps young people have agency in a digital world. “Lots of the decisions that affect our lives are already being taken by AI systems,” Colligan said, and with computer science literacy, people have “the ability to challenge those automated decisions.”
5. The kids who learn to code will shape the future. They’ll get to decide what technologies to build and how to build them, Colligan said.

Hadi Partovi, the CEO and founder of Code.org, agreed that the value of computer science isn’t just economic. It’s also about “equipping students with the foundation to navigate an increasingly digital world,” he wrote in a LinkedIn blog post. These skills, he said, matter even for students who don’t pursue tech careers.

“Computer science teaches problem-solving, data literacy, ethical decision-making and how to design complex systems,” Partovi wrote. “It empowers students not just to use technology but to understand and shape it.”

With her worried students, York said it’s her job as a teacher to reassure them that their foundational skills are still necessary, that AI can’t do anything on its own, that they still need to guide the tools.

“By teaching those foundational things, you’re able to use the tools better,” York said.

Computer science education should evolve with emerging technologies

If foundational computer science skills are even more valuable in a world increasingly powered by AI, then does the way teachers teach them need to change? Yes, according to experts.

“There is a new paradigm of computing in the world, which is this probabilistic, data-driven model, and that needs to be integrated into computer science classes,” said Colligan.

The Computer Science Teachers Association this year released its AI learning priorities: All students should understand how AI technologies work and where they might be used, the association asserted; students should be able to use and critically evaluate AI systems, including their societal impacts and ethical considerations; students should be able to create and not just consume AI technologies responsibly; and students should be innovative and persistent in solving problems with AI.

Some computer science teachers are already teaching about and modeling AI use with their students. York, for instance, allows her students to use large language models for brainstorming, to troubleshoot bugs in their code, or to help them get unstuck in a problem.

“It replaced the coding ducks,” York said. “It’s a method in computer science classes where you put a rubber duck in front of the student, and they talk through their problem to the duck. The intention is that, when you talk to a duck and you explain your problem, you kind of figure out what you want to say and what you want to do.”

The rise of generative AI in K-12 could also mean that educators need to rethink their assignments and assessments, said Allen Antoine, the director of computer science education strategy for the Texas Advanced Computing Center at the University of Texas at Austin.

“You need to do small tweaks of your lesson design,” Antoine said. “You can’t just roll out the same lesson you’ve been doing in CS for the last 20 years. Keep the same learning objective. Understand that the students need to learn this thing when they walk out. But let’s add some AI to have that discussion, to get them hooked into the assignment but also to help them think about how that assignment has changed now that they have access to these 21st century tools.”

But computer science education and AI literacy shouldn’t just be confined to computer science classes, experts said.

“All young people need to be introduced to what AI systems are, how they’re built, their potential, limitations and so on,” Colligan said. “The advent of AI technologies is opening up many more opportunities across the economy for kids who understand computers and computer science to be able to change the world for the better.”

What educators need in order to prepare students for what’s next

The challenge in making AI literacy and computer science cross-curricular is not new in education: Districts need more funding to provide teachers with the resources they need to teach AI literacy and other computer science skills, and educators need dedicated time to attend professional development opportunities, experts said.

“There are a lot of smart people across the nation who are developing different projects, different teacher professional development ideas,” Antoine said. “But there has to be some kind of a commitment from the top down to say that it’s important.”

The Trump administration has made AI in education a focus area: President Donald Trump, in April, signed an executive order that called for infusing AI throughout K-12 education. The U.S. Department of Education, in July, added advancing the use of AI in education as one of its proposed priorities for discretionary grant programs. And in August, first lady Melania Trump launched the Presidential AI Challenge for students and teachers to solve problems in their schools and communities with the help of AI.

The Trump administration’s AI push comes amid its substantial cuts to K-12 education and research.

Still, Antoine said he’s “optimistic that really good things are going to come from the new focus on AI.”