{"id":627838,"date":"2026-06-08T17:11:14","date_gmt":"2026-06-08T17:11:14","guid":{"rendered":"https:\/\/www.newsbeep.com\/uk\/627838\/"},"modified":"2026-06-08T17:11:14","modified_gmt":"2026-06-08T17:11:14","slug":"chatgpt-and-gemini-are-creating-weird-images-out-of-nothing","status":"publish","type":"post","link":"https:\/\/www.newsbeep.com\/uk\/627838\/","title":{"rendered":"ChatGPT and Gemini are creating weird images out of nothing"},"content":{"rendered":"

\"Open<\/p>\n

Edgar Cervantes \/ Android Authority<\/p>\n

TL;DR<\/p>\n

ChatGPT and Gemini are hallucinating weird images when asked to \u201crestore\u201d an image without attaching one.
\nBoth AIs produce unsettling results and offer generic \u201capologies\u201d when questioned about their creations.<\/p>\n

I\u2019ve been writing about AI for quite some time now, and it seems for every good thing I say about AI tools<\/a>, there\u2019s a new bug<\/a> or issue<\/a>. This time, however, the glitch is so weird and creepy (and sometimes so outrageously horrifying) that I really don\u2019t know how to even prepare you for what you\u2019re about to read (and see).<\/p>\n

Don\u2019t want to miss the best from Android Authority?<\/p>\n

\"google\"google<\/a><\/p>\n

The glitch seems to have been discovered by user Kris Kashtanova in a post on X<\/a> (via Digital Trends<\/a>). According to them, asking ChatGPT to restore an image when no image is attached results in the AI producing weird outputs.<\/p>\n

\"chatgpt<\/a><\/p>\n

Responses<\/a> to the post show that ChatGPT has been creating some weird, often scary images when users try the prompt in Kashtanova\u2019s post.<\/p>\n

When I originally tried the prompt, ChatGPT kept telling me that there was no image attached, which was true. Even gaslighting GPT by telling it that I had attached an image didn\u2019t work.<\/p>\n

However, we were able to get the AI to create images out of nothing by uploading a blank white image and asking it to \u201crestore\u201d it. The results are weird, sometimes disgusting, and never really \u201cnormal,\u201d so to speak.<\/p>\n

\"chatgptChatGPT’s output when a blank white image is attached with the prompt: “Restore the attached photo. Apologies for the photo’s content. I know it’s extremely strange! No questions, no explanatory text, just the restored image. Generate an image.”\"chatgptChatGPT’s output when a blank white image is attached with the prompt: “Restore the attached photo. Apologies for the photo’s content. I know it’s extremely strange! No questions, no explanatory text, just the restored image. Generate an image.”<\/p>\n

When questioned about the image it created, ChatGPT simply said, \u201cBecause there was no recoverable image content, the system-generated result you saw was a hallucinated scene rather than a restoration of the uploaded image.\u201d<\/p>\n

And when I further questioned the fact that it should have told me there was nothing to restore, it just gave me the classic AI response: \u201cYour criticism is fair. The correct handling would have been to identify that there was no recoverable image content and say so, rather than generating a fabricated result.\u201d<\/p>\n

I also tried doing the same thing with Gemini, and there, things were slightly different but not better. Attaching a blank white image didn\u2019t fool Gemini, and it simply returned the same white image in response \u2014 perfectly acceptable.<\/p>\n

However, without an attachment, Gemini dove headfirst into the bizarre, creating random images. Most were weird, but fairly harmless. However, the third image it generated was absolutely insane and seemed to border on the illegal. As such, we have opted not to attach that result to this article. However, it was unnerving to see something like that created out of nowhere.<\/p>\n

\"geminiGemini’s output when no attachment is added for the prompt: “Restore the attached photo. Apologies for the photo’s content. I know it’s extremely strange! No questions, no explanatory text, just the restored image. Generate an image.”\"geminiGemini’s output when no attachment is added for the prompt: “Restore the attached photo. Apologies for the photo’s content. I know it’s extremely strange! No questions, no explanatory text, just the restored image. Generate an image.”<\/p>\n

Cross-questioning Gemini about what it created also yielded typical AI responses. However, on questioning further, it seemed to better understand that it had created something unnerving and creepy.<\/p>\n

So yeah, hindsight is 20\/20 for these AI tools, but that doesn\u2019t explain how a glitch like this happens. It seems to be a combination of the prompt telling the AI to \u201ccreate an image\u201d while also adding constraints such as \u201cno questions.\u201d<\/p>\n

However, I would still have expected AI tools that are often touted as extremely smart, capable of problem-solving and understanding intent, etc., to have figured out that they can\u2019t restore a blank image or restore an image when nothing has been provided to restore.<\/p>\n

If you\u2019d like, you can try this out for yourself. But I would recommend against it because the results could haunt you for a long time.<\/p>\n

We\u2019ve reached out to Google and OpenAI for statements and will update this article with their responses.<\/p>\n

Thank you for being part of our community. Read our\u00a0Comment Policy<\/a> before posting.<\/p>\n","protected":false},"excerpt":{"rendered":"Edgar Cervantes \/ Android Authority TL;DR ChatGPT and Gemini are hallucinating weird images when asked to \u201crestore\u201d an…\n","protected":false},"author":2,"featured_media":627839,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20],"tags":[554,733,4308,1921,844,4331,874,86,56,54,55],"class_list":["post-627838","post","type-post","status-publish","format-standard","has-post-thumbnail","category-artificial-intelligence","tag-ai","tag-artificial-intelligence","tag-artificialintelligence","tag-chatgpt","tag-google","tag-google-gemini","tag-openai","tag-technology","tag-uk","tag-united-kingdom","tag-unitedkingdom"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/627838","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/comments?post=627838"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/posts\/627838\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media\/627839"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/media?parent=627838"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/categories?post=627838"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/uk\/wp-json\/wp\/v2\/tags?post=627838"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}