For years, teaching children about artificial intelligence was a matter of individual schools deciding it mattered. That has changed. England's Curriculum and Assessment Review has put AI and digital literacy at the centre of what schools will be expected to teach, and the direction of travel is now settled. This post explains what the review said, what it means in practice, and why treating it as a 2028 problem is the wrong read.
What the review actually said
The Curriculum and Assessment Review, led by Professor Becky Francis, published its final report in November 2025, and the government responded the same day, accepting its direction. On computing, the conclusion was blunt: the current curriculum is too narrow for a digital world, and pupils are not developing the digital literacy they need for everyday life or work.
The review also rejected a common assumption, that children pick these skills up on their own. They do not, and the rise of generative AI has made the gap more urgent rather than less. Digital literacy, as the review defines it, is not being handy with a device. It is the knowledge, behaviours and confidence to use technology safely and effectively, and to make informed, critical judgements about its risks and impact. That is a description of AI literacy in all but name.
Computing becomes the home of AI literacy
The review confirmed that computing will remain the main subject responsible for teaching digital literacy, critical thinking and foundational AI concepts, with clearer expectations about what pupils should learn at each key stage. Related strands sit alongside it: responsible technology use and online safety in RSHE, and the critical evaluation of online content, including spotting misinformation, in a citizenship curriculum that is becoming compulsory and reaches into primary school.
For schools, the takeaway is that AI literacy is no longer an enrichment activity that lives or dies on one teacher's enthusiasm. It is becoming a defined, expected outcome, with computing as its anchor.
A broader Computing GCSE, and a runway to prepare
Two further changes matter. The narrow GCSE in Computer Science is to be replaced with a broader Computing GCSE that includes data and the use of AI, and the government will explore a new post-16 qualification in data science and AI. BCS, the professional body for IT, has been asked to lead the drafting of the updated computing curriculum.
On timing, the honest picture is a runway rather than a cliff edge. The revised national curriculum is expected to be published in spring 2027, with first teaching from September 2028. Nothing is mandatory tomorrow.
Why many schools aren't waiting
It would be easy to read "2028" and file this under "later". That would be a mistake, for three reasons.
The first is that pupils are not waiting. They already use AI daily, with or without guidance, so the risk the curriculum is responding to is present now, not in three years. The second is that schools which start early build something their peers won't have: staff who are confident teaching it, and a track record they can point to. The third is reputational. Parents increasingly ask how a school is handling AI, and "we're waiting for the curriculum" is a weaker answer than "here is what we already do".
Starting now also means arriving at 2028 with the hard part already solved, rather than beginning from scratch under a deadline.
Where to start
A school does not need the final programmes of study to make a sensible start. The direction is already clear enough: teach pupils to understand, question and use AI safely across KS2 and KS3, as a structured progression rather than a one-off, and keep a record of what they have actually learnt so the school can show its work.
That last part, evidencing learning rather than just delivering content, is the piece schools most often wish they had built in from the start, and it is what we designed Curio to handle. Schools starting this autumn get their first six weeks free, which is a low-risk way to get ahead of the changes. If it would help to talk it through, we'd be glad to.