Question
Why does AI add things I didn't ask for?
Because a model fills every gap your description leaves with what is typical for that kind of image, from props and people to stray text. If you don't say what the background, props, people or text should be, it chooses for you. Describe the scene fully, and name what should be absent when it matters.
Image models learn from very large collections of pictures, so a "kitchen" arrives with typical kitchen objects and a "poster" often arrives with lettering. The less you specify, the more the model supplies from those patterns.
Describing what you do want is usually more reliable than listing what you don't: "an empty white counter" gives a clearer target than "no clutter". Where something specific must not appear, say so plainly in the same request, and keep busy scenes simpler, since crowded scenes give more room for stray objects.
If one unwanted object appears in an otherwise good image, edit rather than regenerate: ask for that object to be removed and say what to keep.
In ImageQuarry you can make that removal as a plain-language request on the version you like, and the cleaned image is kept as a new version alongside the original.
Source: editorial guidance from ImageQuarry, checked October 20, 2026, with OpenAI's image generation guide for how edits work.
Related questions
Last reviewed October 20, 2026 by ImageQuarry
Create an image. Edit it with words. Keep refining.
Remove what you didn't ask for
Keep the image you like and ask ImageQuarry to take out the one thing that doesn't belong.
Start creatingAlready have an account? Sign in