AI was undoubtedly the theme of Dassault Systèmes’ 3DExperience Conference 2026, which took place this week in Houston, Texas. Dassault announced a new collaboration with Nvidia, revealed new AI assistants, and even showed off how Solidworks can use itself.
AI loomed even larger on the show floor. Exhibitors at the 3DExperience World Playground were keen to show off how they’re using AI, how they could help you use AI, or how their AI tools could revolutionize your work. We went booth-to-booth to find the most interesting AI applications, and here are our top picks.
Design for manufacturability assisted by AI
Generative design developer InfinitForm was demoing its platform that focuses on making manufacturable generative designs. We first reported on the company in 2024 as they secured a $2.5 million investment to launch.
Historically, generative design tools have had a reputation for creating optimized parts that are impossible to actually make using standard manufacturing processes, necessitating complex and expensive 3D printing. InfinitForm, with the help of an AI assistant, generates parts that can easily be machined.
“The way our AI works is more of a copilot, rather than an AI tool that helps with optimization and FEA analysis. All that is just math-based applications and algorithms,” Adam Smith, InfinitForm’s head of business development, h told Engineering.com. “Our AI is more about how do I help a customer get onboarded? It’s a knowledge base. It’s a pre-check before they run the optimization and as a post-check optimization.”
InfinitForm currently integrates with Solidworks and Siemens NX, allowing users to bring designs back into those CAD programs with machinable geometry. The company says that integrations with other CAD software are in the works.
Improving athlete performance through skate sharpening
Dassault Systèmes always highlights a slew of customer use cases at their annual conference, both in the general sessions and on the Playground floor. This year that included pinball machines, toothbrushes, and hockey skate sharpening company Sparx Hockey.
Sparx says over 10 million skate sharpenings a year are done with their sharpening machines, which were designed and tested in Solidworks. The company’s next step is into the world of data. Sparx’s machines have long been collecting performance data and sending it into the cloud, and this year Sparx wants to pair that data with player data being collected by division one hockey and NHL teams.
“We will begin over the next year or two to unlock how does skate preparation impact performance? Right now it’s a lot of subjective testing that goes on, where players go to rinks with different blades,” Russell Layton, founder and CEO of Sparx Hockey, told Engineering.com. “The difference in the blades are so small that the human might not be able to feel it.”
Using artificial intelligence and machine learning, Sparx plans to unravel how variations in the way skates are sharpened affects everything from heart rate to respiration data in order to maximize player performance.
“Right now, we’re just piping all the data into the cloud. We’re using Power BI and things like that to graph and chart all this data and the trends,” Layton said. “Can you find statistical differences in performance, acceleration, top speed, heart rate to number of strides over certain distances?”
Collecting high quality data to train robots
Another 3DExperience user that was heavily highlighted during the conference was bionic hand company Psyonic, which is creating robotic hands for both humans and robots. Over 300 humans are now using the company’s bionic hand as a prosthesis and over 50 robotics companies are using it in their machines. One of those users is NASA, whom Psyonic CEO Aadeel Akhtar said he’d be visiting while in Houston to see the hand in NASA’s bipedal Valkyrie robot.
Psyonic is also directly collaborating with Nvidia on a large scale data collection project, Psyonic’s main tie into the world of AI. The force feedback that the company’s hands provide to humans is data that can be used to train robots on dexterous tasks.
As Akhtar recently said in a LinkedIn post, “This is what real-to-real transfer looks like. Humans performing real tasks in the real world, showing robots how to do it the same way with the same hand. The lack of high quality manipulation data is one of the biggest challenges in robotics, and this collaboration is about building that foundation, improving dexterity for both humans and robots.”
This data is provided by humans performing tasks for robots to copy. Psyonic can collect the data through a phone app that allows users to control the robotic hand by moving their own hand in front of the phone’s camera. 3DExperience World attendees could try this at Psyonic’s booth, and it worked pretty well. It wouldn’t pick up extremely fast movements, but it could easily differentiate fingers and hand motion and command the robot hand to respond quickly. It gave a lot of thumbs ups over the course of the week.
For more on Psyonic, read Let’s get physical: A new prosthetic restores people’s sense of touch.
Making machining more intelligent
The HCL CAMWorks booth highlighted how AI can streamline the machining process Bruce Wiener, CAMWorks’ manager of technical services, explained that the CAM software can learn how your company machines certain kinds of parts, taking into account things like the shape or material.
“For instance, if it’s a tap hole of a certain size, you can teach it what your machine process is for those types of holes,” Wiener told Engineering.com.
Wiener added there may be other ways to extend machining intelligence. AI could enhance feature recognition, a longstanding offering of CAMWorks, to adjust to a user’s preference for how a part is broken down into specific machinable features. CAMWorks ultimately wants to make its software more intelligent to assist machine programmers, providing them answers and information without having to reference help documents.
Creating a new generation of 3D screens
It could be easy to walk by Beijing-based BOE’s booth thinking they were just showing an image of an animated eyeball on a screen. But stop for a moment in front of the display and suddenly that eye pops out of the screen. BOE was showcasing their 16K resolution 3D displays that do not require glasses.
Related: 3D screens are back, and they’re going pro
BOE’s monitor has a built-in camera that checks your eyes as you walk up to the screen, which sends each eye a different image to make it appear three dimensional. BOE used AI to train their cameras as well as in eye tracking. It’s not a radical use of AI, but it creates a very effective 3D illusion.
Right now the display only works for one person at a time, but Jade Li from BOE’s business development and product planning department says the company is developing a screen to work for multiple simultaneous users as well.
BOE is targeting medical training and diagnosis applications for their initial deployment (hence the eyeball displayed on the screen). The company came to 3DExperience World to explore the potential intersection of their technology with CAD.
“We’re talking about integrating our hardware with Solidworks software. And then designers can really see what they’re designing in 3D format,” Li told Engineering.com. “We still need to work this out, but this is not so hard.”
AI honorable mentions
With AI so prominent on the 3DExperience Playground, we couldn’t get in depth on all of it. There were a lot of other organizations deploying AI in interesting ways. Some worth mentioning are:
- Electronics test and measurement company Keysight showcased their Eggplant software, which uses AI to help automate test creation and test execution. Keysight acquired Eggplant in 2020 and has brought their team underneath the Keysight umbrella. The software uses computer vision to directly “read” screens to pull in information to run tests across many device types, browsers, operating systems, screen resolutions and more.
- In the education section of the Playground was Texas A&M
’s WIRED Underwater Robotics team, showcasing last year’s autonomous robot creation (the incredibly named Swim Shady). The student team uses computer vision tool Roboflow to train the fully autonomous robot, and will even be incorporating a robotic arm onto the robot this year. - Westwood Robotics brought their robots Themis and Bruce to display and wave at crowds throughout the week. On stage during the general session, CEO Xiaoguang Zhang highlighted the company’s “AI augmented humanoid OS which is a modular stack that coordinates whole body stability navigation and safe interaction with consistent behavior,” as well as the robots’ ability to adapt with minimal data inputs.
And that’s a wrap on 3DExperience World 2026. Next year the user conference heads to Nashville, Tennessee. We’ll be watching to see how these AI applications continue to evolve and develop over the next year.
The post 5 Fascinating AI applications at 3DExperience World 2026 appeared first on Engineering.com.
