A (3:22)

Yeah. So let's start with industrial world model.

B (3:26)

Okay.

A (3:28)

Our ambition in fact is to build AI for industry. It's very, very important for US industries. It's at the core of everything we do. And for us AI for interest. Industry rely on three core principles. It should be grounded in science. And this is what we do for more than 40 years. Now we are a scientific company, we deliver modeling technologies, simulation technologies then should be fueled by industry knowledge and it should be sovereign by design from the underlying infrastructure up to the model themselves. So how is it different from a generative AI? I think a classic generative AI learned the dynamics of the world from the observation and the perception of the world. Let's imagine they can see a video of a plane. They can predict if the plane will take off, if it will fly. But in fact they don't really know why, because they don't have the scientific explanation and the scientific foundation right to understand that. And obviously a plane does not fly by accident. So in fact our industry world model principles, they understand how things works, they really understand the scientific foundation. They include the scientific physics laws of the world, the physics, engineering rules, chemistry, material science, etc. And they combine the multi scale, multidiscipline modeling and simulation technologies we provide with AI and the technology we are delivering. Our industry wide models rely on three technical pillars. First, industrial knowledge. Here we are talking about the standards, regulation, the processes from the different industry we serve. Yeah, and we embed the real world engineering rules so the AI will understand and will speak the language of the industry, the jargon of the industry. Then the virtual the world understanding. World industrial understanding. Here we are delivering an ecosystem of specialized industrial AI models which operate on our virtual twins. So the virtual and real representation of the product you deliver.

B (6:03)

Right, Right.

A (6:04)

And this integrates the structure and the physics behavior. So combined with our DAS system modeling and simulation technologies and solvers, this is how we can ensure that the AI will be grounded in science. And last is the industrial reasoning and generation. And this is where the agentic choreography take place and activating the industrial knowledge and the world presentation to perform the experience based resume. And so about virtual companions. Now if in fact, if the industry world model provides intelligence, the virtual companion turns that intelligence into action. What we mean with virtual Companions is virtual Companions are your coworker. They understand your intent, of course, but they will reason with industry world models to orchestrate execute action in context of your business, of your industry, so they will comply with regulation with your KPIs, etc. And they will protect your most precious iPad, of course. And something important, we don't want to replace people, we want to augment people. We want to free time to people to innovate and solve problems. So a few months ago we introduced three virtual Companions. Ora, the business expert, Leo, the engineer who solve complex engineering challenges, and Mari, the scientist who bring deep scientific expertise.

B (7:41)

So when you're designing and deploying the virtual companions, and if we think about sort of a workforce, a virtual workforce of companions that, as you said, aren't replacing human workers, but working side by side with us in an environment like in a manufacturing environment or industrial environment, where I think of my work in content, creating content, podcasting and writing. And if an LLM hallucinates, then hopefully I catch it and I can make the correction or maybe it inspires me to something. If a system hallucinates in an industrial environment, the consequences could be much more dire. So how do you build trust into these systems so that the people who are designing and deploying and working in these environments feel confident working alongside the virtual Companions?

A (8:35)

In fact, a fixosome, the foundation for trust in our system is the scientific foundation, scientific background, then the human in the loop, because at the end, human is accountable and remain in the loop. And the choreography will pause when human have to take decision at the critical milestone of the execution and something very important we deliver. And I think which is unique is what we call iplm, IP Lifecycle management. Okay. And where we enforce the lineage, auditability, traceability of all the interaction of AI. So we are able to know that your content has been modified through which workflow, using which what kind of models, etc. Etc. And we provide the source of trust to understand how your virtual companion behave with your content.

B (9:40)

So Nvidia is bringing technologies, open models, OMniverse Accelerated Computing, AI physics libraries, all these technologies into the stack. How do technologies like these help enable more capable and more secure agentic workflows?

A (9:58)

Yeah. So Nvidia technologies in fact infused in every layer of our architecture from Nvidia AI with AI factories for GPUs and computing infrastructure to Nvidia AI CUDA X Libraries, Omniverse technologies to accelerate AI training, inference and simulation. Regarding Nvidia AI and Agentic, we focus on our partnership with Nvidia on three axes Understanding, reasoning and execution. Understanding we integrate Nvidia NIMS models into our Outscale Kubernetes platform. Outscale is our iis. It's a brand from Dassault System and we are huge fan on Nims because it's super easy to deploy and perfect.

B (10:51)

Always glad to hear it.

A (10:52)

All our team are in love with this. Awesome.

B (10:54)

Love to hear it.

A (10:56)

So we leverage Nvidia open models for Multimodality, Riva, Parse, VLM and with Parse we improve for example by 30% our document injection and throughput plus also some industry specific models such as Bionemo for our virtual companion Marie the scientist about reasoning. Now we leverage NE3 super and the reasoning performance for Ora, Leo and Mari have been improved by 20% without specific optimization. And this is thanks to the collaboration with Nvidia. We shared our industrial use case and benchmark and so we were able to iterate together and to optimize the model and the integration and then about execution. With Nvidia we are continuously improving the agentic execution, leveraging the recent announcement of aiq, Blueprint and Deep Agent and we are also interested in prototyping the recent announcement of nimuclo of course and we are exploring Dynamo to optimize the GPU utilization and Nemo Agilent toolkits for the optimization of our agentic workflows.

B (12:25)

Can you speak a little bit to the partnership? You've mentioned it as you've been talking, which kind of how it got started and more kind of what it means to Dassault and what it enables you to do.

A (12:37)

In fact, for over 25 years now, as you said, Dasystem and Nvidia have redefined what is possible together. Moving from accelerating pixels to accelerating compute and computing and now to accelerating industrial AI. And so back in back in 20, back in back in 2000, from acceleration of visualization of Catia V5, our flagship brand and app, leveraging Nvidia GPUs to accelerating computing for Simulia Abacus and Xflow, our simulation brand with CUDA and of course GPUs to accelerating and optimization rendering with Ira RTX and now with a DLSS. And so this year we are opening a new chapter in this story with AI and the combining Nvidia technologies within our 3 dexformance platform to deliver industrial AI platform to our customers.

B (13:44)

I want to ask you about open and proprietary models and running a hybrid model. My understanding is that Dassault runs hybrid models quite a bit. Can you speak a little bit to kind of the pros and cons of each. And why you go with the hybrid model so often.

A (14:03)

Yeah, you're right. We have a hybrid approach. Of course we build our own models. Yes. But we want to rely on the best in class frontier model provided by Nvidia, such as the Nemotron, our optimized model by Nvidia and available through nims, which as I said before, enable a seamless deployment. It's super easy. Or we have also a partnership with other model providers such as Mistral. In fact, we select our models and our partners based on the performance of the model, of course, but also about the sovereignty and the regulation constraint.

B (14:42)

Okay.

A (14:43)

Because we operate worldwide, we have a customer in all industry and many customers in regulated or very sensitive industries.

B (14:53)

Sure.

A (14:54)

So we have to comply with our own regulation and all the auditability.

B (14:59)

Right.

A (15:02)

And so from that we also want to calibrate the model with the customer knowledge. So we inject the industry knowledge through fine tuning or rag depending of the use case. Sure. But more generally we believe in open standards and so we embrace and we support open standards such as MCP or agent to agent. In fact it, it empower our agency platform to leverage third party industrial system and enable in fact interoperable or cross system agent choreographies.

B (15:40)

I want to ask if we can dig in a little bit to a specific use case to kind of get a flavor for some of the things your customers are doing. Maybe if there's an example that comes to mind you could speak to that really illustrates the use of the virtual companions and the Dassault platform.

A (15:57)

I think one super cool example I think is a LEO mechanical designer. We showcase this live, this new virtual companion in our 3D Experience World Conference last February with Jensen attending to this conference. And so here you give Leo a 3D scan or 2D drawing or a mesh of a part. It will activate the industry world model for design, orchestrate the AI model and the modeling and simulation solvers. And it will perform a multi tier planning, enabling, evaluating in fact the mechanical interface of the part. Find the physics, the kinematics and the design rules. And at the end it will generate the optimized design physics aware manufacturable manufacturer ready. And it will do it right the first time.

B (16:59)

Amazing.

A (17:00)

It's a very super example. I think it really illustrates our transformation from a SaaS to an agent as a service platform. And in fact with that we are giving to our million of designer the power to innovate faster. Yeah, but it's not just about speed, it's about reliability and trust. And because you know that your design works because it is born from science, from physics and is augmented with your industry knowledge.

B (17:32)

Right. That change that you referenced from a SaaS company to an agent, as a service company, kind of from a philosophical standpoint, I guess, or an emotional standpoint, does it feel natural? Is it a big shift? Is it just kind of part of the way of doing things to keep innovating and delivering for your customers? And so it's just kind of the natural progression of things. How do you think about it?

A (18:00)

That's really about, in fact, when we rise of AI, we think ourselves, what is the deep impact of AI in what we do, in what we deliver, what will be the new experience for the user, what will be the new technology we will see the cloud code, etc. What if you apply such transformation to our industrial software? In fact. So it came from that, in fact. Really. And so this is a lot of discussion and brainstorming at the system. In fact, we don't want to add AI on top of what we do. We want to put AI at the core. And this is why we are working with Nvidia on the default topics.

B (18:49)

What's a typical way to get started? What's the first project that a customer might typically undertake to get started with Virtual Companions and working with them?

A (18:59)

I think you should start from your core business and your core challenge. In fact, course, this is where you will have attention from your teams. This is where you have your knowledge, your deep knowledge and your deep know how. And this is where, this is how you know to measure the real impact of your AI and agentic transformation.

B (19:25)

Right.

A (19:26)

And we have an example of connecting to LEO mechanical design. We are working with Nayar and Naya is one of our customers working with us on Virtual Companion. And what they are doing to do is they recreate the virtual twin of existing aircraft. It means that they are creating thousands of parts without access to the original design. So basically they disassemble the aircraft and recreate virtually piece by piece.

B (19:56)

Right. Wow.

A (19:57)

So of course with leo, you can imagine how it changed their life. Automatically generating the 3D part from their multiple sources.

B (20:07)

That's incredible. So like everything else in technology, in AI now, virtual twins, virtual companions, simulation just accelerating, advancing so quickly. And obviously agentic frameworks and models are developing just as quickly, if not faster. What's next? What's on the horizon for Dassault systems? What are the kinds of things you're thinking about? And then if you're game. To take it a step further, where do you think Agentix Systems and the idea of virtual co workers is headed?

A (20:43)

Okay, first, I think the system strategy is fully aligned with the recent Nvidia announcement about Nemo, clo, aiq, all the Agentix stuff and the. The rise in fact of the long running autonomous agent. And we fully agree on the, on the associated industrial challenges, security compliance, etc. And tomorrow, our virtual companion, Oralio and Mari, we believe they will stay awake and they will continuously monitor your factory, your project execution, your, Your supply, your supply chain in real time. And they will proactively optimize it, optimize the virtual twin without being prompt by a human. So it will create in fact, I think a closed loop autonomy. And because of our industry world models are grounded in physics, I think the agents can use the virtual twin as a gym to train themselves so they can run in fact millions of simulation or design experimentation and present to you, to the human, to the engineers, the proven solution and you just have at the end to validate. And from that the virtual twin in fact become a self evolving asset that gets smarter day after day. In fact.

B (22:05)

Nicola, there's so much going on for listeners who want to learn more, want to learn more about the 3Dexperience platform, about Dassault's work, with everything we've talked about, virtual companions and industry world models, where's a good place to go? The Dassault website, social media, are there research papers? Where can listeners go to learn more?

A (22:27)

Mainly on the Dassau system website3ds.com or on our LinkedIn page where we are communicating more and more on AI. Thanks also to the Nvidia collaboration. We are posting more and more about. About what we are doing. So. Yeah, perfect. Yeah, that's free. And connect with us.

B (22:47)

Excellent. Well Nicola, again congratulations on all the work and thank you for the years of collaboration with Nvidia. Thank you and best of luck in everything you're doing.

A (22:57)

Yeah, thank you to Nvidia, to the team, the incredible team.