On March 26, OpenAI said that the initial release of Images in ChatGPT mainly focuses on image creation and can be used by users of Plus, Pro, Team or free subscription packages.<\/p>\n
Strengths of the new creation image<\/h2>\n
On social networks, many people tried it out and expressed surprise about the new tool.<\/p>\n The Verge <\/em>Citing OpenAI spokesperson Taya Christianson, the free version will have limited features, but will still outperform Dall-E.<\/p>\n \u201cThe new feature is a huge step forward compared to the previous model,\u201d said lead researcher Gabriel Goh, adding that his team used the multimodal platform GPT-4o, one of OpenAI's most powerful language models, for ChatGPT's imaging capabilities.<\/p>\n According to Goh, a notable improvement in ChatGPT's image generation capabilities using GPT-4o is called \"Binding\" - a term that refers to the degree to which the AI \u200b\u200bimage generator maintains precise associations between properties and objects. For example, given the prompt of a blue star plus a red triangle, a pattern with poor alignment produces only the red star without the triangle. Goh says most image models struggle with this, often mixing up colors and shapes if receiving multiple requests at once.<\/p>\n \u201cThe new visualization engine with Binding can accurately bind attributes for 15-20 objects without confusion, thereby demonstrating a significant improvement in accuracy and reliability,\u201d Goh said.<\/p>\n The image generator on ChatGPT has also improved the display of text in images, helping to create text that is more coherent and not \"distorted\".<\/p>\n Additionally, the new tool uses an automatic regression method, which creates images sequentially from left to right and top to bottom similar to writing text, instead of the diffusion modeling technique used by most imagers.<\/p>\n \u201cThis is an iterative process that takes many months to perfect,\u201d Goh emphasized. He added that while not perfect, the image creation capabilities on ChatGPT \"have reached a point where the quality of the products created is usable.\"<\/p>\n In the new feature demo, OpenAI presents several examples that show ChatGPT's seamless imaging capabilities, such as a scientific diagram of a Newtonian prism experiment with precisely labeled color components;<\/p>\n However, compared to other models, Images in ChatGPT takes longer to create images.<\/p>\n \u201cWe will definitely improve the latency, but the existing ability to generate images, the quality of the images it produces can really make up for the extra seconds of waiting,\u201d Shannon wrote on the blog.<\/p>\n Regarding the risk of creating fake, nude photos..., Shannon said Images in ChatGPT has strong protection features, preventing pornographic deepfake content and rejecting \"fraudulent\" requests, but did not mention details. The generated image also integrates C2PA standard metadata to mark it as created by AI, which can be looked up with tools for detection.<\/p>\n \u201cOf course, no system is perfect, but we continually improve our protections,\u201d Shannon added.<\/p>\n Bao Lam <\/strong>(according to\u00a0The Verge, TechCrunch)<\/em><\/p>","protected":false},"excerpt":{"rendered":" ChatGPT n\u00e2ng c\u1ea5p kh\u1ea3 n\u0103ng t\u1ea1o \u1ea3nh OpenAI t\u00edch h\u1ee3p ch\u1ee9c n\u0103ng Images in ChatGPT, cho ph\u00e9p ng\u01b0\u1eddi d\u00f9ng t\u1ea1o \u1ea3nh tr\u1ef1c ti\u1ebfp trong chatbot th\u00f4ng qua s\u1ee9c m\u1ea1nh c\u1ee7a m\u00f4 h\u00ecnh GPT-4o. Ng\u00e0y 26\/3, OpenAI cho bi\u1ebft b\u1ea3n ph\u00e1t h\u00e0nh ban \u0111\u1ea7u c\u1ee7a Images in ChatGPT ch\u1ec9 y\u1ebfu t\u1eadp trung v\u00e0o t\u1ea1o h\u00ecnh \u1ea3nh […]<\/p>\n","protected":false},"author":1,"featured_media":22300,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"wpai_generated_summary":"","footnotes":""},"categories":[174],"tags":[179,181,199],"class_list":["post-22183","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-cong-nghe","tag-ai","tag-chatgpt","tag-openai"],"acf":[],"_links":{"self":[{"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/posts\/22183","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/comments?post=22183"}],"version-history":[{"count":2,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/posts\/22183\/revisions"}],"predecessor-version":[{"id":23149,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/posts\/22183\/revisions\/23149"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/media\/22300"}],"wp:attachment":[{"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/media?parent=22183"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/categories?post=22183"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/gudjob.net\/en\/wp-json\/wp\/v2\/tags?post=22183"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"|\"](\"https:\/\/gudjob.net\/wp-content\/uploads\/2026\/08\/1-1-6.webp\")
Limitations and feedback from users<\/h2>\n
![\"|\"](\"https:\/\/gudjob.net\/wp-content\/uploads\/2026\/08\/2-1.webp\")