Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124
Physical Address
304 North Cardinal St.
Dorchester Center, MA 02124

Claude Interactive charts, graphs and visuals can be created directly in chat using a new feature Anthropy announced Thursday. The new tool, which is currently in beta (so expect bugs), is not an AI image generator. It’s not a Nano banana A competitor, but something completely different. In fact, you really need to see the feature to get an idea of how interesting and clever it is.
I tried out the new feature when I asked Claude to show me how to change the tire, and it produced a comprehensive, interactive step-by-step visualization that left me confident in my ability to handle the task. It wasn’t just a cool trick (although it was cool); It was a better way to show the process than the typical block of text from an AI-powered chatbot.
These new visuals have a specific purpose: to give users a better understanding of the topic they are discussing with Claude. In some cases, a large block of text can be replaced more clearly with a visual, which is what Anthropic, the maker of Claude, is trying to do here. The system identifies when visuals may be the most appropriate response in a conversation so that information can be more easily understood.
Among the more interesting examples shared by Anthropic are a step-by-step tutorial for creating a “cool” paper airplane and an interactive periodic table that lets you click on individual elements to get more information.
When Claude decides that a visual element is the most appropriate response – or when the user explicitly requests it – he creates interactive HTML and SVG files on the spot, built from scratch. Anthropic likens this feature to giving Claude a whiteboard upon request.
This feature is part of a broader push by Anthropic to make Claude a more dynamic AI tool that can choose the best way to provide an answer. Even without a traditional image generator, the capability can be more impactful, providing visualizations when they are most useful.
The feature will be turned on by default, so there is no need to click into settings to enable it. Plus, it’s available on all plan types, which means you won’t have a paywall to access it. It is currently only available for web and desktop and may be available on mobile in the future.
With simple guidance, Claude gave me a step-by-step tutorial for changing a tire on a car I didn’t own.
To see how this feature works, I asked Claude to show me how to change the frame visually. Within seconds, an interactive illustration of the seven-step process was produced, complete with the tools required and an explanation of the key steps. The illustrations weren’t very detailed, but they clearly got the idea across.
For fun, I also asked Claude to create an interactive visual representation of the facade of a house from a book I was reading. Not only was this done accurately, but almost everything was clickable, displaying bits of information when choosing a window or door. The feature has a real “cool” factor, and you can download the illustrations you create or save them as artefacts for later use in Cloud.