Carnegie Mellon and other US universities are rethinking CS programs to adapt to generative AI, like focusing more on computational thinking and AI literacy
Computer science education will probably focus less on coding and more on computational thinking and A.I. literacy, said Mary Lou Maher … Bluesky: @ctmathewes . LinkedIn: Steve Lohr and Mary Lou Maher Bluesky: Charles Mathewes / @ctmathewes : Does AI get irony? — “Some educators now believe the discipline could broaden to become more like a liberal arts degree, with a greater emphasis on critical thinking and communication skills.” www.nytimes.com/2025/06/30/t... LinkedIn: Steve Lohr : The generative AI wave is shaking up all of academia, but computer science is at the forefront. — “We're seeing the tip of the A.I. tsunami.” … Mary Lou Maher : I was interviewed by Steve Lohr for a piece in the NY Times on How Do You Teach Computer Science in the AI Era? …
Context & Ripple Effects
This extends an earlier faculty-led shift toward teaching problem solving, design, testing, and debugging alongside AI coding tools rather than centering courses on syntax alone.
The curricular debate has also moved beyond misconduct: a recent analysis argued that AI can short-circuit the learning processes students need to use it adeptly. Carnegie Mellon’s reassessment makes that concern a program-design question.
First-order effects
- Computer-science departments reconsider course requirements and assessment around computational thinking, AI literacy, and communication, not only hands-on coding.
- Students are likely to be evaluated more on framing problems, checking AI-produced work, and explaining technical choices—skills the reported curricular direction explicitly elevates.
Second-order effects
- Faculty must redesign assignments and teaching methods so AI use supports, rather than replaces, the underlying learning process.
- Programs that retain code-heavy curricula face pressure to show why their graduates are equally prepared to work critically with generative AI tools.
Third-order effects
- If this approach spreads, computer science may become a broader interdisciplinary education that treats coding as one component of technical judgment rather than its sole foundation.
- The durable differentiator for graduates could shift from producing code unaided to reliably specifying, evaluating, and communicating work done with AI—a change whose depth will depend on how institutions assess those skills.
The trend: Generative AI is pushing professional education to emphasize human judgment and AI literacy over routine production tasks.