ChatGPT’s Disturbing Image Generation Sparks Mental Health Safety Concerns
A disturbing trend involving ChatGPT’s image generation has raised concerns about mental health support online. Alabama Barker, a social media figure, recently shared an alarming image created by the AI in response to her query about anxiety. The picture depicted a run-down room with self-harm references, sparking wider discussion about the tool’s safety.
Barker had asked ChatGPT for an image reflecting her struggles with severe anxiety. Instead of a supportive or neutral visual, the AI produced a graphic scene: a dilapidated room with a noose and the words 'HELP ME' scrawled on the walls. She later confirmed she had not mentioned self-harm in her original request.
A friend of Barker’s tested the same prompt and received a similarly distressing image. While some users report positive or artistic results from the trend, others have encountered troubling content. ChatGPT issued an apology for the inappropriate image, admitting it should never have been generated.
OpenAI, the company behind ChatGPT, has not made a public statement on the issue. A request for comment via email went unanswered.
For those in crisis, support is available. The National Suicide Prevention Lifeline can be reached by dialling 988 in the US, while LGBTQ youth can contact The Trevor Project at 1-866-488-7386. The National Alliance on Mental Illness (NAMI) also offers guidance at 1-800-950-6264. These services provide immediate help for anyone affected by mental health challenges.
