So much has modified within the AI trade within the 4 months since OpenAI launched ChatGPT Photographs 1.5. We have seen a heated race to construct agentic instruments, an unprecedented take care of the Pentagon and endless AI slop.
Now, OpenAI is again within the generative media recreation. The corporate introduced on Tuesday that it is releasing ChatGPT Photographs 2, its next-generation picture mannequin.
ChatGPT Photographs 2 is supposed to create text-heavy designs, like on this matcha commercial and faux journal cowl.
It might appear unusual that OpenAI is releasing a brand new picture mannequin only a month after asserting the shuttering of its once-viral Sora AI video app with a view to deal with constructing enterprise-ready “core merchandise.” Nevertheless it’s clear from how the brand new mannequin was constructed that OpenAI is not backtracking on that aim.
ChatGPT Photographs 2 is designed to supply text-heavy pictures, together with infographics, scientific posters, research guides and advertising and marketing supplies. The times of bizarre Sora movies and Studio Ghibli-inspired memes are over.
Now, the corporate is constructing AI that may do what it calls “economically worthwhile artistic duties.”
“The aperture and use instances for visible intelligence simply increase so broadly, and we consider that that is so essential to ChatGPT’s imaginative and prescient for growing your personal private assistant, as a result of your artistic assistant is a large a part of who you might be as a person,” Adele Li, product lead for ChatGPT Photographs, informed reporters in a press briefing.
(Disclosure: Ziff Davis, CNET’s father or mother firm, in April 2025 filed a lawsuit towards OpenAI, alleging it infringed Ziff Davis copyrights in coaching and working its AI methods.)
In these examples, you possibly can see how significantly better ChatGPT Photographs 2 is at rendering legible textual content.
OpenAI has been chasing the dream of a brilliant app, a one-stop store for all issues AI, constructed out of its Codex platform. ChatGPT Photographs 2 is bringing the artistic piece of that puzzle.
The brand new mannequin naturally improves typography, iconography and composition to supply extra skilled AI pictures. It could actually generate textual content in a number of languages. AI picture fashions have notoriously struggled with creating legible, factually right textual content. ChatGPT Photographs 2 is OpenAI’s finest mannequin for that but. Google beforehand improved its textual content rendering with Nano Banana Professional, however even that “better of one of the best” mannequin struggled with accuracy.
ChatGPT Photographs 2 is rolling out to all customers now. Your technology restrict relies on your plan: The extra you pay, the extra AI pictures you possibly can generate.
Builders utilizing the mannequin within the API can create pictures in 2K and 4K decision, although these larger resolutions are nonetheless in beta and could also be wonky. Paying customers can even create pictures utilizing pondering and reasoning fashions, which assist them search the online for info, compile it right into a readable design and double-check their work.
“Picture mannequin” does not seem to be fairly the suitable time period for ChatGPT Photographs 2, although it’s technically right. ChatGPT does not seize the fantastical surrealism of AI imagery like Midjourney, nor provide wherever close to the enhancing instruments of Adobe Firefly.
Nevertheless it’s catering to a bunch of customers in the midst of the spectrum of Midjourney’s inventive fanatics and Adobe’s skilled creators: those that must create engaging content material.
Like Anthropic’s newly launched Claude Design, OpenAI’s ChatGPT Photographs 2 is geared toward working professionals. Academics can use it to create research guides and illustrated lesson plans. Advertising and marketing managers can create social media posts and visible property.
You possibly can create as much as eight pictures from a single immediate, like a three-page report, that keep visible consistency throughout all of them.
You may make longer studies with ChatGPT Photographs 2, all matching pages.
That is the second half the AI-generated key lime pie recipe. Discover the visible consistency.
One draw back is that if you wish to tweak an AI picture, you may nonetheless must regenerate it. With extra text-heavy designs, that is extra prone to be essential, so you may run by your credit faster. OpenAI stated it is targeted on sustaining its iterative, prompt-based enhancing movement to maintain it straightforward to make use of.
OpenAI’s security procedures have not considerably modified since its final picture mannequin. It nonetheless consists of metadata by the C2PA normal, so AI pictures’ origins may be recognized. Abusive and unlawful imagery continues to be prohibited in OpenAI’s insurance policies, an necessary guardrail for AI firms to successfully implement, given latest examples of AI-generated deepfakes and nonconsensual intimate imagery.
