Gemini gets 3D simulations users can actually manipulate
The visual shows the difference between a static diagram and a simulation the user can adjust.📷 AI-generated / Tech&Space
- ★Google says the new Gemini feature generates interactive simulations directly in chat
- ★Examples include a Moon orbit, double pendulum, and Doppler effect
- ★The rollout is global for Gemini users who select the Pro model, but not for Education and Workspace accounts
Google has added interactive 3D models and simulations directly inside the Gemini app. Instead of a static image, a user can receive an object that rotates, a model that pauses, or a system whose variables change through sliders and numeric inputs.
In its official announcement, Google describes a Moon-orbiting-Earth example with adjustable initial velocity and gravity strength. The Verge saw a similar model with an orbit-speed slider, a toggle to hide the orbital path, a pause button, zoom, and rotation.
That is a real change from earlier multimodal answers. A static diagram can explain the shape of a phenomenon, but not what happens when a parameter changes. A simulation, even a simple one, forces the user to see cause and effect: more velocity, a different path; different gravity, a different system.
Availability is the important nuance. Google says the feature is rolling out globally to Gemini app users who select the Pro model in the prompt bar, but the note attached to the announcement says it is not currently available for Education and Workspace accounts. That is awkward because schools and universities are exactly where these models could have the clearest value.
Google is moving AI answers from static diagrams into interactive models, but usefulness will depend on accuracy, education access, and whether the result can escape the chat box.
The mid-article visual separates education potential from professional simulation precision.📷 AI-generated / Tech&Space
The competitive context is not subtle. OpenAI introduced interactive visual explanations in March for more than 70 math and science concepts, while Anthropic has pushed Claude toward charts and other visual answers. Google's response is broader in form: not only graphs, but 3D models and physics scenes.
An educational simulation, however, is not the same thing as a professional engineering tool. If Gemini shows a double pendulum or the Doppler effect, the user needs to know whether it is a qualitative explanation or a numerically reliable model. Without explicit units, cited equations, and export options, it remains a demonstration inside a chat.
The best case is better understanding. A learner who manipulates a lens, orbit, or wave can grasp a relationship that prose might blur. The worst case is false precision: a smooth animation that looks like a lab but represents an approximation no one has validated.
Gemini 3D is therefore important without being self-proving. Google has added a new kind of answer, not a new standard of truth. The next test is whether this interactive layer gains cited formulas, reliable behavior, school access, and an API path for tools that need to live beyond one chat window.
