Quick answer
Neither answer to this question is honest on its own. AI helps teenage creativity when it's used as an idea-partner - generating options, offering an alternative angle, doing rough groundwork the teenager then judges and reshapes. It hurts creativity when it's used as an idea-replacement - producing a finished piece the teenager accepts and submits without reworking it. The tool is identical in both cases; the habit is what differs. Because AI can now generate a plausible draft in seconds, the creative skills that matter most have shifted from raw production to taste - knowing what's actually good - and editing - having the judgement to make it better. Those are the two things worth watching for and actively building.
Why this question doesn't have a simple yes or no
Parents understandably want a clean verdict: is AI good or bad for creativity? The honest research picture doesn't offer one, because "using AI" covers wildly different behaviours. RAND's American Youth Panel found 67% of students said AI use for schoolwork harms critical thinking - a finding that tracks closely with creative thinking, since both depend on wrestling with a problem rather than accepting the first plausible answer. Gerlich's 2025 study in Societies, of 666 participants, found heavy AI use strongly associated with cognitive offloading - letting the machine do the thinking - and offloading associated with weaker critical thinking, most pronounced in 17- to 25-year-olds.
But none of that research says creative use of AI is harmful. It says offloading is harmful, and creative work is exactly where the line between "using a tool" and "offloading the thinking" gets blurry fastest, because a finished poem, image or story looks the same to an outside observer regardless of how much actual creative work went into it.
The idea-partner versus idea-replacement test
This is the distinction that actually predicts the outcome.
An idea-partner relationship looks like this: a teenager stuck on a story asks AI for five possible directions, rejects four, and writes their own version of the fifth. A design student asks AI to generate rough mood-board images, then builds their actual concept by combining and rejecting elements from what came back. In both cases, AI expanded the option space; the teenager still did the choosing, the judging and the finishing.
An idea-replacement relationship looks like this: a teenager asks AI to write the story and submits it. A student asks AI to generate the final artwork and hands it in as their own. Here, AI didn't expand anything - it substituted for the creative act entirely, and the teenager's own taste and judgement never got exercised.
| Signal | Idea-partner (helps) | Idea-replacement (hurts) |
|---|---|---|
| What AI produces | Raw options, rough drafts, alternative angles | The finished piece |
| What the teenager does next | Selects, edits, combines, adds their own voice | Accepts and submits |
| Can they explain a specific choice? | Yes - they made it | Often not - the tool made it |
| Effect over time | Taste and editing skill grow | Taste and editing skill atrophy |
Why taste and editing are the new creative skills
Generation used to be the hard part of creative work. A first draft, a rough sketch, a demo melody all took real effort to produce, which meant producing anything was itself evidence of some creative capability. AI has made raw generation nearly free. Anyone can now produce a plausible draft, a competent-looking image, or an inoffensive tune in seconds.
That shift moves the scarce skill somewhere else: to judgement about which of fifty generated options is actually worth keeping, and to the editing skill required to turn a rough AI output into something with real voice and intention behind it. This is the same pattern described more broadly in the durable skills AI cannot replace - taste is on that list precisely because it requires responsibility and exposure that a model cannot substitute for. A teenager who spends time building taste - looking at a lot of work, getting honest critique, learning to say "this option is better and here's why" - is training a skill AI generation makes more valuable, not less.
How to tell which habit your teenager has formed
You don't need to inspect every piece of creative work your teenager produces. A few honest questions do most of the work.
- Ask about a specific choice. "Why that ending?" or "why this image over the others?" A teenager who made the choice can usually answer in a sentence. One who didn't often shrugs.
- Ask to see the discarded options. Real creative work using AI as a partner tends to leave a trail of rejected drafts. Idea-replacement usually doesn't, because there was only ever one version.
- Watch what happens without the tool. Ask your teenager to sketch, draft or outline something small with no AI involved. If the instinct to generate ideas is still there, the underlying skill is intact.
Common mistakes parents make on this question
- Banning AI from creative work entirely, which removes a genuinely useful brainstorming tool and does nothing to teach the judgement that actually matters.
- Praising polish over process, which rewards impressive-looking AI-replacement work over rougher, genuinely original AI-partnered work.
- Assuming creativity is fixed talent, when taste and editing are trainable skills that respond directly to exposure and honest feedback.
- Not asking any questions at all, and mistaking a finished-looking piece for evidence that real creative thinking happened.
The recommendation: stop asking whether AI is good or bad for creativity and start watching for the habit - idea-partner or idea-replacement - because that's what actually determines the outcome. Ask your teenager to explain their choices, keep AI in the brainstorming and rough-draft lane rather than the finished-product lane, and treat taste and editing as skills worth deliberately building, not talents you either have or don't. Handled this way, AI becomes one more source of raw material for a creative mind that still does the deciding - part of the broader picture in AI education for teenagers in Australia.
Frequently asked questions
Related insights
Written by
Lachlan Matheson
Lachlan Matheson writes for Edison AI Insights on practical AI adoption, capability and the everyday habits that turn new tools into real advantage.
Published by Edison AI Academy · About the academy
Learn AI the Edison way, with judgement built in.
Edison AI Academy teaches ambitious Australian students to think, build, and lead with AI through structured, project-based, responsible education.
