The cloud has always focused on text and code, but a few days ago, it introduced image-generation capabilities — ways to capture AI-generated graphics, mock-ups, slideshows and other similar content. A set of tools is called Cloud design.
Then ChatGPT, already heavily invested in AI image generation, gets a significant update. Images 2.0. Promises were made of “step changes” in accuracy, consistency, and following instruction.
So what can you do with these AI tools now that you couldn’t before? And how do the capabilities compare?
Cloud design is focused on business and enterprise
Cloud Design is available to Cloud subscribers, and is intended, in Anthropic’s words, for “visual work such as designs, prototypes, slides, one-pagers and more.” It won’t produce pictures of cats riding skateboards for you, but it does will do Pull together a project slideshow, or mock-up of an iPhone app.
This continues the cloud’s focus on business and enterprise: as well as coding with the cloud, teams can work on prototypes and put together pitch decks. It may sound a little dry, but the new capabilities are really versatile and include spinning, interactive globes, such as Demo video.
Cloud Design will produce an editable slide deck for you.
Credit: Lifehacker
To get started, you can get Cloud to create visuals from an existing codebase, load existing images and documents to use as starting points, or simply write a text prompt. You need to go special Cloud Design landing pageSeparate from the main chatbot interface, it lets you choose how you want your workflow to run.
I decided to create a slide deck showing the value of Lifehacker and give the AI a few screenshots to work on for a style idea. Claude then asked me a few questions about what I wanted, including the mix of text and images and how long the slideshow should be before getting to work. You’ll see the AI ”think” and work through the steps to create graphics in real time.
When the finished work was presented, it was impressively polished—and the cloud design gives you everything you need to export your work elsewhere. One of the most useful features is the way you can tweak visuals after they’ve been created – on my Slide Deck, I was able to tweak accent colors, fonts, and slide density with just a few clicks.
A mock-up of a possible Lifehacker News app.
Credit: Lifehacker
You can also request edits through further prompts, and even draw visuals to indicate what should happen next. Moving on to the iPhone mock-up of a possible Lifehacker news app, Cloud Design did a great job here too, barring one or two minor graphical glitches: the app design looked very Lifehacker-y, and I could request edits just by drawing on the visuals and typing what I wanted to change.
It’s all smart, professional and easy to get around; I can see a lot of companies using cloud design along with cloud code. For individuals, it looks like a useful way to put together ideas for designing just about anything, including slide decks (though the AI tools inside apps like PowerPoint and Google Slides might suit you better).
ChatGPT Images 2.0 focuses on consumer as well as business use
As for ChatGPT and its Images 2.0 upgrade, it is more generalized and focuses on consumer as well as business use. OpenAI says prompt instructions are now followed more closely, end results are more accurate and consistent, and text rendering is is further improved. Functions can be more complex, and images also appear more “deliberately designed”.
What do you think so far?
Creating images is as easy as ever: just click Create an image And explain what you want to see in the prompt box. People are making it complicated where is waldo Images, Infographics from scientific papersAnd Mock magazine cover; I was able to whip up a quick comic strip about Lifehacker in minutes.
Lifehacker Comic-Note that the desk is completely rearranged and the coffee gets hotter.
Credit: Lifehacker/chatgpt
ChatGPT also proved capable of mocking up two Lifehacker magazine covers of varying quality: they certainly look realistic enough, and there are no glaring errors or typos, but at the same time they have that generic feel that comes with many AI images. You can say that this cover “averages out” all the magazine covers taken in ChatGPT’s training data.
Choose your favorite Lifehacker magazine cover.
Credit: Lifehacker/chatgpt
You can’t create slide decks or anything complex in the cloud design with ChatGPT Images 2.0. You could theoretically create single slides and app mock-ups, but there are far more limitations in terms of consistency and editing what’s on screen later — ChatGPT is more about one-off moments of AI art.
The new Images 2.0 model is also better at fetching real information from the web, so you can put together a cartoon map of Middle Earth (though it’s copyright-savvy not to exactly copy Tolkien’s work), or create an informational diagram about upcoming sports tournaments. I ran both of those tests, and ChatGPT came up with impressive visual results.
Rather than ChatGPT, the closest comparison to Cloud Design is probably the Gemini AI tools available through Google Docs, Sheets, and Slides, which I Wrote about earlier. While you can’t yet create full presentations, you can create slides from simple cues and load in other content as references for design.
As with Cloud Design, you can use follow-up prompts to refine certain aspects of a slideshow or document design without starting over from scratch. And your finished work can be exported to a variety of formats, including PDF and Microsoft Office-compatible file types.
