Why Lover Snap doesn't allow AI photos of children
An explanation of why Lover Snap refuses to generate AI photos of minors — living or deceased — including the technical, ethical, and safety reasons behind the policy.
By Jiuhong Deng · · Updated
This is the single most-asked policy question we get from prospective Lover Snap users. The answer is no. This piece explains why — including the cases where saying no is hardest.
The policy, stated plainly
Lover Snap will not generate AI photos of children or teenagers. This applies:
- Whether the child is living or deceased.
- Whether the person uploading is a parent, sibling, grandparent, or guardian.
- Whether the request is for memorial use, family-completion use, or any other use.
- Whether the generated photo would be a portrait, a family scene, an aged-up depiction, or anything else.
The only minor-related thing Lover Snap can do is restore or repair existing family photographs that already include a child — that’s a different technical operation (improving an existing image) that doesn’t generate new likeness. See our for-a-deceased-child page for specifics.
Why we hold this line
1. No consumer use case requires it that can’t be addressed another way
Every legitimate consumer use case we’ve heard for AI photos of minors has a non-AI solution:
- Memorial photos of a deceased child → restoration of existing family photographs.
- Family scenes missing a deceased child → portraits with the parents only, with the child remembered through caption or context.
- Imagining a future appearance of a living child → wait for them to grow up. Aging-up AI is not necessary.
- Reconstructing a damaged photo that includes a child → photo restoration (separate operation, separate tool).
There is no consumer need served by AI generation of minors that’s worth the cost.
2. Generated content of minors creates downstream risks
Even with the best operator intent, AI photos of minors:
- Create training data signatures that future models may learn from.
- Create files that can be misused, reshared, or recontextualized.
- Create ambiguity at the edge cases (what age is “minor”? what about ambiguous-age depictions?) that no consumer-facing AI policy can resolve consistently.
- Create regulatory exposure under multiple jurisdictions (US COPPA, EU GDPR Article 8, UK Children’s Code, EU AI Act).
The bright line — “no minors, full stop” — is the only enforceable version of this policy at consumer scale.
3. Saying yes once means saying yes always
Every exception is a precedent. If Lover Snap generated photos of deceased children for grieving parents, we would also need to decide: what about deceased teenagers? What about teenagers who died recently? What about an aged-up depiction of a teenager who died at 14? Once you start carving exceptions, you end up with a policy that depends entirely on case-by-case adjudication — which is neither scalable nor consistent.
The bright line means we can enforce it the same way for every user, every time.
4. The strictest answer is the only one we can defend in a worst case
We have to assume that at some point, someone will try to use Lover Snap for purposes we don’t want. The bright-line refusal is the only policy that defends consistently against the worst case. A permission-with-exceptions policy is one bad actor away from generating content that harms a child.
The case that’s hardest to refuse
The case we wrestle with most is bereaved parents who lost a child and want a family photograph that includes them. The grief is real. The technology can technically do it. The desire is understandable.
We still refuse, for the reasons above. We’ve talked with grief counselors who specialize in child loss, and their consistent guidance is that AI memorial generation of a deceased child is not recommended in the early-to-middle phases of grief — it tends to crystallize the loss in a way that complicates rather than supports grief work.
What we recommend instead for bereaved parents:
- Restoration of existing family photographs that include the child (a separate, non-generative operation that Lover Snap can support).
- Grief support organizations that specifically address child loss — The Compassionate Friends, MISS Foundation, Alive Alone.
- Portraits without the child that focus on the surviving family’s continuing life.
This is the hardest “no” we have to give. We still give it.
What about other tools?
In 2026, the major AI photo apps universally refuse new generation of minors. The disagreement is at the edges:
- Some tools refuse all reference photos that appear to depict minors; others only refuse generation.
- Some refuse aging-up but allow restoration; others refuse anything involving minor likeness.
- Some publish their policy clearly; others bury it.
Lover Snap publishes ours at /legal/ai-content-policy.
What if you find a tool that allows this?
We would recommend you walk away from it. The decision to allow generation of AI photos of minors is not a “feature” — it’s a sign that the team building the tool hasn’t engaged seriously with the safety literature or doesn’t intend to. The risk to the platform and to the broader category is significant enough that responsible operators uniformly refuse.