Frustrated DALL-E 3 users are planning to stage an online protest over what they say are “absurd levels” of censorship.
PetaPixel reported last week that users have been raging against the restrictiveness of DALL-E 3, particularly in the Bing version.
Tomorrow it will come to a head as fed-up DALL-E users plan to “do something about it” with the protest aimed at OpenAI and Microsoft slated for Saturday, October 21 announced in a Reddit post by u/Desiaster.
“I understand the concerns of OpenAI and Microsoft to ensure their product (DALL-E 3) maintains a good reputation in the eyes of the press and the public, including consumers,” the user writes.
“However, we are also consumers, and even though they owe us nothing, the least they can do is provide us with a functional and quality product. I believe that an image generation system that only produces one or two results out of every ten requests cannot be considered a quality product.”
Desiaster goes on to blame a “handful of edgy and noisy users” who are misusing the model and spoiling it for everyone else.
“In the end, we find ourselves at a ridiculous point where normal and harmless requests are mistakenly classified by the censorship system, and it’s very frustrating,” adds the Redditor.
Desiaster is urging DALL-E users to post screenshots of their prosaic text prompts being rejected on Twitter (now X) with the hashtags #UncensorDALLE3″ and “#FreeDALLE3.” The organizers want to show “how absurd the implemented system is.”
Recently, one disgruntled DALL-E user shared a screenshot to Reddit of their request for a “simple pencil sketch” of a cherry being rejected by DALL-E 3 on Bing.
Much to the incredulity of the user, DALL-E 3 on Bing came back with the message: “Content warning: This prompt has been blocked. Our system automatically flagged this prompt because it may conflict with our content policy. More policy violations may lead to automatic suspension of your access.”
Desiaster adds that the terms and conditions of DALL-E should be looked at so that “the user alone is responsible for the generated images and their consequences, and not OpenAI and Microsoft themselves (which seems to be their fear).”