Back to blog

The 2026 CSTA Standards Are Out — Computer Science Is Bigger Than Coding, and AI Runs Through All of It

November 15, 2026 (4d ago) · 5 min read

The 2026 CSTA Standards Are Out — Computer Science Is Bigger Than Coding, and AI Runs Through All of It

Every few years, a document comes out that quietly resets what our subject is supposed to be. This summer, the Computer Science Teachers Association (CSTA) published the 2026 CSTA PK–12 Computer Science Standards — the first full revision since 2017, and the first one written in a world where generative AI is in every classroom whether we invited it or not.

I teach ICT, Computer Science and Robotics in Bangkok, so US standards are a reference for me, not a rulebook. But CSTA's framework is one of the most widely used references in CS education anywhere, and this revision says some things plainly that a lot of us have been saying in staffrooms for two years. It is worth your weekend reading time — and it is free to download, print, share and adapt for non-commercial use with attribution.

First, a definition worth stealing

The standards open by defining what computer science is — and, just as usefully, what it is not:

  • CS is not using technology tools like word processors, slide decks, or generative AI tools. Those appear in every subject.
  • CS is not limited to coding. Programming is essential, but the discipline is broader.
  • CS is not tied to one career path.
  • And my favourite line in the whole document: CS is not being replaced by AI — CS is what powers AI.

That last point is the answer to the question parents and students keep asking us: why learn to code when AI can code? Because somebody has to understand the systems doing the coding — how they are built, what data they learned from, where they fail, and what they should never be trusted with. That understanding is computer science.

How the new standards are built

The foundational standards rest on three layers:

Five concepts organise the content: (1) Algorithms & Design, (2) Programming, (3) Data & Analysis, (4) Systems & Security, and (5) Computing & Society. Here is the design decision I find most important: artificial intelligence is not a sixth concept. CSTA identified AI as a priority during the revision, then deliberately distributed AI content across all five concepts instead of isolating it as a separate topic. AI shows up when you study data, when you study systems, when you study society — because that is where it actually lives.

Twelve practices describe what students do, grouped into four categories: Ethics & Social Responsibility, Inclusive Collaboration, Computational Thinking, and Human-Centered Design. Notice that only one of the four is purely technical.

Seven dispositions — habits of mind — are named as outcomes in their own right: creativity, critical thinking, curiosity, persistence, reflectiveness, resourcefulness, and a sense of belonging. A student who can code but quits at the first error, never reflects, and feels CS is "not for people like me" has not met these standards, however tidy their syntax is.

The progression is coherent across schooling: a combined pre-kindergarten/kindergarten band, grade-by-grade expectations for grades 1–5, a single middle school band, and a foundational high school band.

The part secondary teachers should not skip

On top of the foundational standards sit High School Specialty Standards: two levels (Specialty I and II) across six areas — Artificial Intelligence, Cybersecurity, Data Science, Game Development, Physical Computing, and Software Development. Specialty II is explicitly aimed at preparing students for postsecondary coursework and industry certifications.

As a robotics teacher, the Physical Computing pathway matters to me: it gives sensor-and-microcontroller work (my students build with Arduino and LEGO robotics) a recognised place in a standards framework, as a progression rather than an enrichment club. There are also X+CS standards for weaving foundational CS into other subjects like biology, journalism and the arts — a formal invitation to collaborate across departments.

What I am doing with it

Four concrete moves, in case they are useful to you:

1. Auditing my curriculum against the five concepts. Most school CS programmes, mine included, are heavy on Programming and light on Data & Analysis and Systems & Security. Mapping each of my units to the five concepts shows the gaps in one page — and gives me a defensible basis for fixing them.

2. Stopping the "AI unit." If AI is woven through CSTA's concepts, it should be woven through my lessons: a data lesson that asks where training data comes from, a security lesson that covers AI-generated phishing, a society lesson on bias. One special week in October will not do it anymore.

3. Assessing dispositions on purpose. Persistence and reflectiveness are now named outcomes, so I can plan for them: debugging journals, "show me your second attempt" tasks, and rubrics that credit a well-reasoned failed approach. What gets assessed gets taken seriously — by students and by me.

4. Using the specialty areas to design pathways, not just lessons. For older students, the six specialty areas are a menu for electives and clubs with a real progression behind them, instead of one-off projects that end when the term does.

The honest caveat

These are US standards, built for US policy contexts. If you teach an international or national curriculum elsewhere, treat them the way I do: as the clearest current statement of what the field thinks CS education is for, and a checklist to interrogate your own programme with — not as a document to adopt wholesale. The questions it asks travel well, even where the answers will differ.

The through-line of the whole revision is a healthy one for our subject: computer science is a human-centred, ethical, creative discipline that happens to involve code — and in the age of AI, that broader definition is not a luxury. It is the whole job.

If you mapped your curriculum against the five concepts tomorrow, which one would come out thinnest?


Sources: Computer Science Teachers Association, 2026 CSTA PK–12 Computer Science Standards — standards overview (DOI: 10.1145/3820482); how the standards were developed.

Title image: original clay artwork created for this blog.

LINE