The honest answer is that the floor sits somewhere around 7–8 for most kids, and the ceiling is wherever they want to take it. The floor is set by reading fluency, attention span, and the cognitive shift Jean Piaget called concrete operational: the point where a child reasons reliably about things in front of them. The signals that say your kid is ready live in your kid, not on a chart. Four of them are worth watching for.
There is no magic age.
Parents ask this question for a reason. Every screen-time decision before age 7 felt high-stakes. AI feels higher-stakes still. So “what age is right?” is shorthand for something more honest: what is the youngest age I can introduce this without doing damage, and what is the oldest age I can wait until without my kid falling behind their peers?
The short version is that the floor is around 7–8 for most kids, and there is no real ceiling. The American Academy of Pediatrics has spent the last decade revising its screen-time guidance to focus less on hours and more on what the screen is doing for the child.1 A passive feed at age 5 is one thing. A creative tool at age 8 is a different thing, and the research does not lump them together. So the question is not “how much screen” but “is the kid building something or being delivered something?”
The floor exists because three things have to be in place before a kid can use an AI tool productively: enough reading fluency to parse a chat message, enough attention span to hold a project across a sitting, and enough abstraction to reason about a thing they cannot touch. All three usually come online together, between roughly age 7 and age 8. For some kids it is earlier. For some it is later. The age is a guide, the readiness is the actual signal.
Why earlier is hard.
Before about 7, most kids are in what Piaget called the pre-operational stage.2 They can use symbols and language, but they have trouble with logical operations. Reversibility (if A turns into B, can B turn back into A?) is shaky. Conservation (does the amount of water stay the same when you pour it into a taller glass?) is unreliable. Probability and causation at any complex level are essentially out of reach.
This matters for AI specifically because AI tools require the kid to hold three ideas at once. The proposal the AI made. The state of the project before the proposal. The state of the project after, if the proposal is kept. A pre-operational kid can see the after-state. They have a hard time holding the before-state and the proposal at the same time, which means they cannot meaningfully evaluate the change.
You can watch this happen. Show a 5-year-old a wizard proposing two character designs for a game. They will pick one because it is on the right side of the screen, or because it is purple. They are not weighing the two options against the project they are making. The cognitive equipment is not there yet, and asking for it produces frustration rather than learning.
Modern developmental psychology has refined Piaget’s ages but kept the rough shape. Annette Karmiloff-Smith’s neuroconstructivist work argues that development is more continuous and more domain-specific than the stage labels suggest, with kids doing some advanced reasoning early in domains they care about and lagging in domains they do not.3 A 6-year-old obsessed with dinosaurs can reason about taxonomy that a typical 8-year-old cannot. The general shift, though, between “reasoning about things in front of me” and “reasoning about systems and rules” tracks the age range Piaget identified.
Why ages 7 to 8 tends to work.
Around age 7, three things usually arrive in close succession. Reading fluency reaches the point where a kid can decode chat messages without help. Attention span lengthens enough to sit with a single project for twenty to thirty minutes. And Piaget’s concrete operational reasoning comes online: the child starts reliably reasoning about objects, comparing states, and noticing when something changed.2
The concrete operational stage is the floor for AI tools because it lets a kid do the basic loop: see what was, see what the AI proposed, decide if the after-state is better. That is the smallest unit of learning the studio supports. Below the floor, the loop breaks because the kid cannot hold the before-state. At and above the floor, the loop works, and the kid starts accumulating the iteration habit other posts on this journal describe.
The concrete operational stage is the floor because it lets a kid do the basic loop: see what was, see what the AI proposed, decide if the after-state is better. Below the floor the loop breaks. At and above it, the loop teaches. Piaget’s concrete operational stage, applied to AI tools
Vygotsky adds a second consideration.4 The zone of proximal development is the gap between what a kid can do alone and what they can do with a knowledgeable partner. The zone exists at every age, but its character changes. A 6-year-old’s zone is narrow and concrete: with help, they can do things a little harder than what they can do alone. A 9-year-old’s zone is wider and more abstract: with help, they can reach into systems and rules that they could not touch unaided. AI tools are most useful when the zone is wide enough that the AI’s proposals can stretch the kid’s reach. Below 7 or 8 the zone is usually too narrow to make AI scaffolding pay off.
This is also why the American Academy of Pediatrics’ current guidance focuses on co-use and content selection rather than age cutoffs.1 A 7-year-old building with an AI partner alongside a parent is in a different regime than a 7-year-old left alone with the same tool. The presence of a more knowledgeable other widens the zone, which is exactly what Vygotsky predicted.
Four readiness signals in your kid.
The age range is a guide. The signals that say your kid is ready live in your kid. Four of them are worth watching for, and they tend to arrive in this rough order between ages 6 and 9.
1. Independent reading. Can your kid read a sentence on a screen without help, get the gist, and act on it? Not perfect comprehension. Just enough to follow a chat exchange. If they can read the back of a cereal box and tell you what it says, they are usually there.
This is the most reliable single signal because the studio’s chat surface is the kid’s primary interface. Inkie writes; the kid reads; the kid writes back. Without independent reading the kid is dependent on a parent narrating every line, which collapses the autonomy that makes the loop work.
2. Asking “why” questions. Not the toddler “but why?” loop. The genuine kind: “why does the dragon breathe fire when I hit the spacebar?” A kid asking real why-questions is showing that they hold mental models and want to update them. That is the engine of the iteration habit.
3. Sustained 20-minute focus. Can your kid sit with a single thing for 20 minutes without bouncing? It does not have to be reading or homework. LEGO counts. A drawing counts. A pretend-play scenario counts. If they can hold attention on a self-directed task for that long, they can hold attention on a project in the studio.
4. Willingness to revise. The strongest signal. Show your kid a drawing they did last week and ask if there is anything they would change. A kid who shrugs and says it is fine is not ready for an AI tool yet. A kid who points at three things and tells you why is. The studio’s loop is built on revision. A kid who already revises their own work casually will take to it. A kid who does not yet will find the workflow alien.
If your kid shows three of the four, the studio is going to work for them. If they show two, give it a few months. If they show all four, you can probably skip the “is my kid ready” worry entirely and just hand them the demo.
How we landed on 8 in the studio.
The studio’s age range is 8 to 14. The floor is 8, not 7 or 6. Three reasons we landed there.
First, the cognitive floor. The four signals above usually consolidate around age 8 for most kids. Setting the floor a year below that, at 7, would catch some kids but produce frustration for the median. Setting it at 9 would miss kids who are ready. Eight is the median floor where the typical kid can run the loop unaided. Mitchel Resnick’s Lifelong Kindergarten group has been arguing for forty years that creative tools for kids should have low floors and high ceilings.5 The lowest floor that still actually teaches is the one to aim at.
Second, parent supervision changes the math. The studio is designed so a parent can sit next to a 6 or 7-year-old and run the chat for them. We do not recommend this for daily use, but it works for the first session, or for siblings paired together. The 8 floor is the floor for independent use. The presence of an older sibling or a parent widens the zone and pulls the floor down.
Third, the artifacts the studio ships are real software, deployed to real URLs, sharable to real audiences. That is heavier weight than a kid-coding sandbox where the artifact lives in a closed system. We chose 8 because it is the age at which most kids can hold the responsibility of having shipped something to the public web, and ask thoughtful questions about it. Below that age, the public-shipping dimension can become more pressure than reward.
The honest summary: 7–8 is the floor for most kids using AI tools that ask them to make decisions. The studio sits a year above that floor to keep the typical first session productive. Your kid’s actual readiness, not their birthday, is the signal that should drive the call. The studio has a 30-day refund window precisely because we know some 8-year-olds are not ready, and some 9-year-olds were ready a year ago.
References
- American Academy of Pediatrics, Council on Communications and Media, “Media and Young Minds,” Pediatrics, Vol. 138, Issue 5, November 2016. Plus the AAP’s ongoing Family Media Plan guidance at healthychildren.org/MediaUsePlan, which focuses on co-use and content over hours.
- Jean Piaget & Bärbel Inhelder, The Psychology of the Child, Basic Books, 1969 (translated from French). Stage definitions including pre-operational and concrete operational. See also Piaget, The Construction of Reality in the Child, Basic Books, 1954.
- Annette Karmiloff-Smith, Beyond Modularity: A Developmental Perspective on Cognitive Science, MIT Press, 1992. The foundational neuroconstructivist argument that development is continuous and domain-specific rather than stage-locked.
- Lev Vygotsky, Mind in Society: The Development of Higher Psychological Processes, Harvard University Press, 1978 (translated from Russian originals from the 1930s). The zone of proximal development is in chapter 6.
- Mitchel Resnick, Lifelong Kindergarten: Cultivating Creativity through Projects, Passion, Peers, and Play, MIT Press, 2017. The low-floor / high-ceiling / wide-walls standard. MIT Media Lab Lifelong Kindergarten group at media.mit.edu/groups/lifelong-kindergarten.