Asger Risom
AI pioneer Geoffrey Hinton is urging students not to abandon computer science.
Amid ongoing debate about the future of tech education, AI pioneer Geoffrey Hinton is urging students not to abandon computer science. Speaking with Business Insider, he said the field still offers enduring value even as artificial intelligence absorbs more routine programming tasks.
His comments come at a moment when students face a tougher job market and experts reassess what skills will matter most in an AI-driven industry.
More than programming
Hinton told Business Insider that misconceptions about the degree persist. “Many people think a CS degree is just programming or something,” he said, arguing that mid-level coding roles are the ones most at risk as AI advances.
He stressed that the discipline teaches broader principles and systems thinking, describing this foundation as the reason a “CS degree will be valuable for quite a long time.”
Other tech leaders cited by Business Insider expressed similar views. OpenAI chairman Bret Taylor called a CS education “extremely valuable,” noting that “there’s a lot more to coding than writing the code.”
Evolving expectations
While broadly supportive, industry figures say curricula must shift. Google’s Android chief Sameer Samat previously told Business Insider that the field should be reframed around “the science … of solving problems.”
UC Berkeley professor Hany Farid added that the most compelling roles for graduates now lie outside Silicon Valley’s major platforms. In an interview with the outlet, he said the biggest opportunities appear “at the intersection of computing and other fields,” listing areas from medical imaging to computational finance and digital humanities.
These perspectives reflect a widening consensus that CS programs should prepare students for interdisciplinary applications rather than narrow software roles.
Advice for younger learners
Hinton also defended the value of early coding education. He said learning to program remains “a good intellectual activity,” comparing it to studying Latin as part of a broad academic background.
“I think it’s very useful to learn to code,” he told Business Insider, even if future tools automate much of the work. He argued that the discipline cultivates reasoning skills that AI cannot replace.
For those aspiring to advanced AI research, Hinton recommended prioritising core analytical abilities. He said knowledge of “math, and some statistics, and some probability theory … linear algebra” will endure even as tools evolve.
Sources: Business Insider
