Most product design conversations basically still happen around a flat screen, which is sort of strange since almost nothing about a real product is actually flat. So when Jim Napolitano and the team at Dassault Systèmes recently took home the BIG Innovation Award with 3DLIVE, they weren't just building another visualization tool. They were, in a way, betting that the next leap in spatial computing in product design would actually come from making the virtual twin feel like an object you can walk around, touch, and pull apart with your hands.
That bet, frankly, seems to have paid off. 3DLIVE, built directly for the Apple Vision Pro, basically lets engineers, executives, surgeons, and consumer research subjects step inside a data-rich version of a product well before it ever exists physically. And the kicker is that it's not just a snapshot. It really is a live feed of the design data, so the model evolves as the product does. That detail might sound minor, yet it tends to be the thing that quietly reframes the entire design review cycle.
What a Virtual Twin Actually Is (and Why It's Not Just a 3D Model)
Ok, so let's get the vocabulary right first. A virtual twin, basically, is a data-fueled version of a product, and at Dassault Systèmes the twin sits at the heart of pretty much everything across their 11 industry verticals. According to McKinsey research on digital twins, companies that operationalize this kind of model tend to see meaningful gains in development speed and product quality, sometimes shaving months off the cycle.
Jim explained it pretty plainly: a virtual twin lets you simulate a wind tunnel test or a crash test before you ever build the physical prototype. You can also tweak the dynamics on the fly until the product is more or less hitting whatever success criteria you set. That alone really changes the math on prototyping, which has historically been one of the biggest line items in product development.
It's also worth flagging that a virtual twin and an AI agent are not the same thing, and every other LinkedIn post tends to use the terms interchangeably these days. The twin, basically, is the data. The agent is the thing that acts on the data. Naturally, both are useful. They just play different roles.
Why Apple Vision Pro Is the Breakthrough Hardware
Here's where things get really interesting. Most of us have tried a previous generation AR or VR headset and walked away with a sore forehead and a vague sense of motion sickness. Past devices, frankly, came with a steep learning curve. You had to figure out the controllers, learn the gestures, and orient yourself in space before you could actually get any work done. According to a Deloitte Tech Trends report, the enterprise XR market has, for years, been gated on exactly that kind of friction.
The Apple Vision Pro, by Jim's account, pretty much collapses that learning curve. Users put the device on and just start interacting with the model naturally. They actually reach over and press buttons. They feel like they're pressing them too. That tactile sense, in a way, is what really unlocks the rest of the application. It's a big reason 3DLIVE works for senior executives and surgeons just as well as for the engineers who've been living inside CAD tools for years.
The Use Cases Are Wilder Than You Might Expect
Cars are obviously the starting point. Many F1 teams already use Dassault Systèmes software somewhere in their stack, so the team built a demo where a full-scale virtual twin of an F1 vehicle sits right in front of you. You can layer aerodynamic simulation data on top of it and watch the airflow respond in real time. You can also grab the front wheel assembly, pull it apart, and rotate the brake caliper around like some kind of real-world Iron Man scene. It's pretty much the kind of analysis engineers used to do with stacks of Excel spreadsheets and sensor logs.
The use cases just keep getting better from there:
- An aviation customer is, for instance, using 3DLIVE to reconfigure passenger cabin layouts virtually, letting reviewers sit inside the aircraft before any retrofit dollars are spent
- A CPG company is, in fact, running consumer research with packaging mock-ups in a fully virtualized retail aisle, so they can study how shoppers actually respond to label and shelf configurations
- A medical device customer is, frankly, using a virtual twin of a living heart to help surgeons train for delicate procedures and even troubleshoot scenarios like a sudden blood pressure drop on the table
That last one is just a little staggering when you actually think about it. A surgeon could, in fact, walk through your specific heart before they walk into the operating room.
Live Data, Real Collaboration, and Faster Decisions
The judges flagged two things that really set 3DLIVE apart. The first is collaboration. Multiple people in different cities can, of course, review the same 3D model together inside Apple Vision Pro and workshop edits in real time. The second is the live data backbone. Most XR product applications have, historically, taken static snapshots of design data at a single moment in time. 3DLIVE, instead, is wired into the actual evolving product data inside Dassault Systèmes software, so what you see today reflects what just got changed yesterday.
What does that change about how teams make decisions? Pretty much everything. Approval cycles compress as reviewers no longer wait days for a fresh export. Consumer research moves earlier in the cycle since the product doesn't actually need to be physically built to get tested. Training scenarios scale once a single virtual twin can spin up thousands of troubleshooting drills. The thread connecting all of these, naturally, is that smart products tend to come from smart decisions made faster, and 3DLIVE is more or less a tool for making those decisions right in front of the actual data.
The Apple and Nvidia Partnerships Quietly Driving It Forward
Two strategic relationships are sort of quietly fueling all of this. The collaboration with Apple has been close enough that Dassault Systèmes engineers spent roughly 18 months working hand in hand with Apple's hardware teams to get the experience right. And the Nvidia partnership, which made headlines when Jensen Huang stood on stage and basically declared that everything is going to be a virtual twin, is plugging in the AI muscle that lets these models simulate even more complex scenarios faster.
It's a useful reminder that even the most innovative products really aren't built alone. The future of spatial computing in product design isn't actually going to come from better headsets alone, frankly. It's going to come from who can wire the virtual twin straight into live data so reviewers can act on it the moment it changes.
Enjoying insights from industry leaders? Subscribe to The Winners' Circle podcast on your favorite podcast player and never miss an episode. Listen and subscribe at bintelligence.com/podcast.