Openaai updated its multimodal AI model Chatgpt 4o on 25 March. It no longer separates the voice model from the image generator. Former imaging model Dal-E is therefore replaced. With a Chatgpt Plus account, it can also be tried in Germany.
Since then, the relevant platform on Reddit and social media has been the lead actor and musician in the style of Japanese cartoon studio, The Simpsons or Studio Ghibali, who is responsible for films like “Princess Mononok”, “Chihiro Ki Trip to the Magic Land”. A simple text -propirt and an uploaded picture is enough to elite the surefire style of many artists. It is a lot of fun with style scenes with musk, remedo, warder. Meanwhile, Ghibali’s co-founder Hoyo Miazaki is an old clip of viral, which AI called the material “insult to life”.

As is well known, the Zegtist in the United States is on a cruel interpretation of freedom of expression, the leading of Elon Musk, and thus here: his image generator, which is integrated into Ki-Chhatbot Grok from XAI, rarely follows general guidelines when the pictures are handled by the copyright-proactive. Openaai is now floating on this wave as well. CEO Sam Altman announced in X that the new Openai image generator allows a higher degree of independence: “We think it is right to keep this intellectual freedom and controls in the hands of users.”
This was our Heise Plus article “Chattgpt 4o read with laxen morality guidelines with image generator”. With a heise plus membership you can read and hear the entire article.
