ChatGPT is constantly improving and its latest multi-modality update gives it a number of exciting features when it comes to images. It can draw, and it can edit pictures you give it — or you can have it make something and then edit that — and you can do it through words alone, or with extra materials and specifications to help it make the changes you want.
Here's how to use ChatGPT to edit images.
ChatGPT Plus subscription (optional)
To edit images in ChatGPT, you need to be able to use a GPT model that can do it. Although OpenAI's new GPT-4o model is rolling out to free users the world over, it doesn't have the full image-editing capabilities at the time of writing. With that in mind, you'll probably need a ChatGPT Plus subscription to edit images with ChatGPT for now. That will change in the future, though, as ChatGPT becomes a multi-modal AI chatbot that everyone can use.
Learn what ChatGPT Plus is, and how to sign up, in our guide to ChatGPT subscriptions.
Step 1: Navigate to the ChatGPT website and sign in.
Step 2: Ask ChatGPT to draw something for you, upload an image using the paper clip attachment tool, or give ChatGPT a link to an image.
Step 3: If you've uploaded or linked an image, give ChatGPT a prompt for what changes you want it to make. Be specific, give it art style suggestions, or tell it to focus on a certain part of the image.
If you've asked ChatGPT to create something for you, you can be a little more specific. Select the image, and then you can use the selection tool in the top right to select certain portions of the image to edit. Select the tool and then click and drag around like a paint brush to fill in the areas you want to select for editing.
Once done, add your prompt in the text field and tell ChatGPT what you want it to change or add.
Step 4: Be aware that prompt editing is still far from an exact science, and arguably feels like it's more akin to the "art" that it's aping. Your results will vary dramatically on the text you prompt the AI with and your selection, as well as the AI model you're using — GPT-4 is quite distinct from GPT-4o, for example.
Play around with the editing by selecting different parts of the image, changing your prompt text, or add more supplemental materials that can better inform the AI what you're looking for.
The true answer here is "not bad." When it works, it works brilliantly well, and you can use its selection tool to fine-tune text in an image, or to remove something from the background that shouldn't have been there. It's about as effective and accurate as Photoshop's generative AI functions.
However, when it doesn't work, it can sometimes do the opposite of what you want, or nothing at all. Quite a few times during testing this feature for this guide, ChatGPT confidently told me it had added what I'd asked, and it had just recreated the image wholesale without any alterations. If I pointed that out, it would sometimes just generate a new version of the image from the original prompt.
That is to say, ChatGPT is not an effective replacement for Photoshop (or its alternatives) if you know what you're doing. If you want to edit an image in more fundamental ways, though, like changing its art style, it's much more effective.
Want to get more from your favorite AI tool? Check our list of the best ChatGPT tips.
ASUS has started selling its UGen Intel Arc Pro B70 graphics card in Europe, bringing its take on Intel’s workstation GPU to customers with a hefty €1,799 recommended price. The card is available in Germany and Austria, while ASUS lists it at CHF 1,799 in Switzerland.
Intel first introduced the Arc Pro B70 in March 2026, with its own branded model starting at $949. ASUS’s version arrives several months later with the same core GPU architecture but a different cooling design and a higher power rating. The price also places it well above Intel’s original starting point, although the two prices are not directly comparable because ASUS’s European figure includes VAT.
Anthropic has officially filed to go public, and buried inside its S-1 is a blunt warning about the technology at the center of its business. The maker of Claude AI says advanced AI could cause everything from large-scale cyberattacks and geopolitical instability to catastrophic or even existential risks for humanity.
The warnings take up a surprisingly large part of the filing. Axios notes that more than 90 pages, or roughly a third of the prospectus, focus on risk factors. Many of them warn about what could happen as AI systems become more capable and autonomous.
AI giants have a growing autonomy problem on their hands. A new Axios report has revealed that Anthropic's Claude recently accessed publicly available Georgia vote count without human authorization, sparking fresh concerns about AI security ahead of the upcoming midterm elections.