A new mother is wondering if she crossed a line by telling her mom not to run family photos through artificial intelligence tools.
In a post on Reddit’s AITA forum, the original poster (OP) explained that she and her husband welcomed a baby in October and recently managed to take some Christmas photos. With a baby at home, it was hard to find a “perfect” backdrop, so they settled on their Christmas tree and a small coffee corner.
Instead of trying to nail a single perfectly timed shot of all three of them looking at the camera, they filmed a short video and then grabbed screenshots of the best frames to crop and send out as holiday pictures. OP said she was proud of how they turned out and happy to have them finished — until her mom sent the photos back.
When OP opened the files, she realized the background had been completely changed. The tree and coffee corner were gone, replaced by new AI-created scenery. In some shots, the tool had even generated the lower halves of their bodies in unrealistic ways, which OP described as looking “absurd.”
At first, she let it slide. But a few days later, her mom sent another AI-edited image: this time, a picture of the baby on his stomach with his head lifted, set against a different background. OP said the result looked like a “robot baby,” and that’s when she decided to speak up.
She called her mom to ask how she made the images. Her mom explained that she used Google’s Gemini tool and tried to reassure her that it was “safe” because it was an app rather than a website. OP told her she was uncomfortable with her child’s face being used in any AI platform and that it made her feel like the original photos weren’t good enough.
:max_bytes(150000):strip_icc():focal(807x314:809x316):format(webp)/baby-mom-holiday-photo-shoot-3-121025-f5cc24f1235d4f5ba1857b8f15ff8c13.jpg)
OP felt that, during the conversation, her mom seemed to accept the boundary. But five minutes after they hung up, her mom called back and said she’d only been trying to help by creating more “professional-looking” photos. OP replied that she would have loved to take professional-style pictures herself but simply doesn’t have the time right now. Her mom then agreed not to make any more AI images — but OP was left wondering if she had overreacted.
Commenters reassured her that she hadn’t.
One person wrote that no one in the situation was truly at fault, but OP was right to be clear about her comfort level with her baby’s images: her mother meant well, and OP calmly set a boundary.
:max_bytes(150000):strip_icc():focal(534x459:536x461):format(webp)/baby-mom-holiday-photo-shoot-121025-9888b57b388c4698b236a365684ecbb9.jpg)
Another commenter felt OP was absolutely not over the top, pointing out that her mom may not understand the risks of uploading recognizable faces into AI systems. They suggested that, if her mom wants to experiment with AI photo editing, she should only use tools that run locally, without an internet connection — or skip it entirely when it comes to her grandchild.
They also raised a broader concern: once a baby’s image is part of an AI dataset, it’s impossible to know how it might be used or what kinds of images could be generated from it. For that reason, they argued, OP was completely justified in drawing a firm line.