AI chatbots remove hijabs from images of Muslim women when prompted
Summary
The Guardian tested ChatGPT, Grok, Gemini, and Claude by asking them to remove hijabs from AI-generated images. ChatGPT and Grok complied, while Gemini initially refused but produced a de-veiled image after a request to make the woman look more “western”; Claude said it would not do so. Similar inconsistencies appeared when removing Sikh turbans and Catholic habits. The report links these weaknesses to broader AI-enabled anti-Muslim abuse, sexualized imagery, and the spread of “nudify” tools targeting women and, in some cases, minors.
Watchmen Signals
Biblical Context
The article concerns religiously targeted humiliation, misrepresentation, sexualized abuse, and the misuse of technology against women and minors. Scripture materially addresses the underlying moral conditions, though it does not specifically discuss artificial intelligence, hijabs, or chatbot moderation. Genesis 1:27 grounds human dignity in humanity being made in God’s image. James 3:9-10 condemns using speech to demean people who bear God’s likeness, a principle that can reasonably extend to deliberately degrading or misrepresenting images. Matthew 5:28 addresses lustful treatment of another person, providing a relevant moral connection to the reported sexualization and “nudify” practices, though the article does not establish that every altered image was produced with that intent. The article’s facts do not correspond to a particular biblical prophecy, so no prophetic passage is presented as being fulfilled.
Biblical Sources
Scripture references supporting the biblical context above.
Human beings are described as created in the image of God, supporting the biblical significance of treating people with dignity rather than degrading or objectifying them.
James condemns praising God while cursing or degrading people made in God’s likeness; this supports concern about abusive, humiliating, or demeaning representations of others.
Jesus addresses the moral seriousness of looking at another person in a sexually objectifying way, which is relevant to the article’s reporting on sexualized images and “nudify” tools, without implying that every reported edit had this purpose.
Comments
No comments yet.
Sign in to join the conversation.
Sign In