Good morning, everyone. So I'm Marie Dumas, I'm the new Investor Relations Director for Dassault Systèmes. Some of you might recognize my voice, from my previous years at the company, and I'm delighted to reconnect with many of you today. Joining from Dassault Systèmes with me are Pascal Daloz, our Chief Executive Officer, and Rouven Bergmann, our Chief Financial Officer. On behalf of the team, I'd like to welcome you to Dassault Systèmes' fourth quarter and full year 2025 earnings presentation. Following the presentation, we will open the floor for questions, first from the room and then from participants joining us online. Later today, we will also host a conference call. Dassault Systèmes' results are prepared in accordance with IFRS. Most of the financial figures are presented on a non-IFRS basis, with revenue growth rates in constant currencies, unless otherwise noted. For an understanding of the differences between IFRS and non-IFRS, please see the reconciliation tables included in our press release. Some of the comments we will make during today's presentation will contain forward-looking statements, which could differ materially from actual results. Please refer to our risk factors in our 2024 Document d'Enregistrement Universel, published on March 18th. With that, I will now hand over to Pascal Daloz. Good morning, everyone. Thank you for joining us today to review the Dassault Systèmes performance for the fourth quarter and the full year 2025. Let's start with an opening comment, which is, you know, at Dassault Systèmes, we do not manage only for the quarter. We build platforms that last for decades. I think 2025 was a year of transition. 2026 is a year of execution. And I think those two years are foundations because they are the year when we prepare the next cycle of growth, scale, and long-term value creations. So let's start with the facts. 2025 was a disappointing year for you, but also for us. We finished at the low end of our objective with 4% growth, excluding foreign exchanges. I think this performance does not meet the standard we set to ourselves as part of the long-term plan, and we own that. That said, we moved the company forward, and I think we made meaningful progress on the priorities that matter the most for the long-term success. What moves forward? First, the 3DEXPERIENCE and the cloud. I think we deliver significant wins and competitive displacement. And in 2026, we will build on this momentum, turning headwinds into tailwinds and deepening our leadership in industrial AI. And I think our ambition is clear: it is to remain the partner of choice for all the industries. Second, Medidata and Centric PLM faced challenges in 2025 and weighed our results. I think we are seeing the early signs of the recovery at Centric. And for Medidata, we are investing for the long term, keeping in mind that Life Sciences is undergoing a fundamental transformation, from inefficient documents-based processes to AI-powered Virtual Twins that redefine how pharma innovates, complies, and operates. So this shift is really structural, and the structural changes usually take time. Third, in 2025, we introduce 3D Universes, a new environment where Virtual Twin and AI converge, connecting the virtual and the real in a seamless, dynamic loop. In 2026, we turn this vision into concrete value. Finally, we remain disciplined on costs while continuing to invest in our future growth. Execution matters, and returns matter as well. Now, as we enter in 2026, we are scaling our transformation plan, ensuring every step we take positions Dassault Systèmes and our customers for sustainable successes. And our transformation is built around three strategic pillars. The first one, the product offering. We are reshaping our portfolio, accelerating towards 3D Universes, and investing where scale matters. Second, the go-to-market. We are strengthening the go-to-market execution with targeted end-to-end engagement, especially in Life Sciences for the top 50 larger accounts and consumer industry for the formulated products. We are also transforming our partner ecosystem to generate demand, not just distribute licenses. To support this, we have strengthened the leadership with a transformation officer and an operating officer focusing on the execution and performance. Third, the business model. This is a very important part. As customers accelerated their adoption to subscription and cloud, we are introducing the annual run rate reporting in 2026. I think this will provide a clear visibility into the health and the momentum of our recurring revenue base. In parallel, we are also evolving beyond the seat-based pricing toward a value-based pricing model for AI-powered solutions. Because we don't just deliver the software, I think we are delivering outcomes, and we will capture a fair share of the value we are creating. This transformation is not just about growth. It's about building a resilient, customer-centric company ready for the generative economy. Now, let's step back a little bit and look at the market realities. You know, every industry we serve, whether it's manufacturing, Life Sciences, Infrastructure & C ities, is under intense pressure: supply chain volatility, rising regulation, aging infrastructure, and an urgent need for breakthrough innovation in all the domains. These are not just constraints. In fact, they are catalysts. And this is exactly where Dassault Systèmes steps in, not only as a vendor but as a strategic partner for many of our customers. We are helping them to turn the complexity into a competitive advantage that lasts for decades. In manufacturing, we see two realities. The traditional sectors face margin pressure and demand uncertainty. But at the same time, defense, High-Tech are bold. They are making bold investments where the complexity and the collaboration are the new normal. And this is where the 3DEXPERIENCE platform is becoming the de facto standard, enabling faster cycles, collaboration at scale, and streamlined operations, reducing program timelines to under 18 months in Transportation & Mobility, delivering between 25%-40% efficiency gains in Aerospace & Defense, cutting errors to one-third to almost half in High-Tech through pre-built simulations. That's what we do. In Life Sciences, the pressure is still intense: tighter regulations, rising R&D costs, and the shift toward personalized therapeutics or precision medicines. And you know, the incremental improvement is no longer enough. I mean, the customers, they need a new operating model. With our end-to-end lab-to-manufacturing solutions, we are really helping them to reduce their operating costs by over 30% while turning the compliance into a competitive advantage. Now, in infrastructures and cities, demand for autonomous and resilient systems is accelerating. You know, data center demand will double by 2030. The nuclear infrastructure requires safe decommissioning in many countries. Cities need resilient by design. And I think our AI-powered Virtual Twins reduce the project timeline by over 25% while ensuring safety and compliances. This is really how we are and we create new markets while addressing the most critical challenges of our time. Now, across every sector, our customers, they prove one thing: we don't just talk about AI. We deliver it. In Transportation & Mobility, as I was saying, innovation pressure has never been higher. Products are more complex. Timelines are shorter. Competition is global. Valeo is a good example of this. You know Valeo. It's a global leader in the automotive technology, from ADAS to electrification systems. I think together, we are pushing the boundaries of the Generative Experiences. Here, AI does not only assist, it co-creates. How we do this? By training the Virtual Twin on our synthetic data, we generate thousands of design alternatives, optimize for performances, costs, compliances before a single prototype is built. This is really the new way of working, turning complexity into opportunity. In Life Sciences, our partnership with Catalyst, I think, shows how an industry can be reinvented. By moving from static documents to data-driven Virtual Twins, Catalyst is really redefining the CRO model. The clinical trials become agile, patient-centric, continuously optimized. You know, this is not about fixing the old model. It's really about building a new one. In infrastructure, with TechnicAtome, I think we are redefining how the next generation nuclear systems are designed and operated. The Virtual Twins connect the entire ecosystem, ensuring the traceability, the compliance-based designs, and more importantly, closing the loop between the virtual and the real world. That year, we did also something extremely important. A year ago, if you remember, we introduced 3D Universes. But what does it mean in practice? 3D Universes are not applications. They are knowledge factories where knowledge is enriched, know-how is scaled, and the results are trusted. And AI is the engine, not a generic AI, not a surface AI, an AI which is grounded in science, engineering, and industry. This is not about large language models. I insist on this because with LLMs, you will never be able to build roads. You will never be able to design humanoids. You will never be able to discover cell therapies. This is what our customers, they do. This is really what we do for them to help them to certify what they do. We are building what we call the Industry World Model, which is the next generation of AI after the large language model. What is important is those models understand how the real world works and how to build it. That's something extremely important. Why so? Because those models, they are built on physics. They are trained on decades of industrial knowledge with continuously validated Virtual Twins. It's explainable. It's certifiable. It's trusted. There is one fundamental reason: because in the physical world, you cannot forgive the mistakes. This is the reason why our partnership with NVIDIA matters. I think together, we are combining the Virtual Twins with AI factories and accelerated computing. But maybe longer than the long explanations. Launch the video, please. Physical AI is here, grounded in science and the laws of physics, resetting the world's largest industries. Dassault Systèmes pioneered Virtual Twins to model and simulate the world. For almost 30 years, NVIDIA has helped bring them to life. Now, Dassault Systèmes is combining NVIDIA AI models, CUDA -X, and Omniverse libraries with its Virtual Twin factories to super-accelerate design and engineering with AI physics, to open new paths to molecule and material discovery, and build complex software-defined robotic factories, whether for manufacturing products or a new type of factory for manufacturing intelligence, AI factories, and new Virtual Companions, AI agents who assist engineers and scientists at every step. The new era of Virtual Twins has arrived, powered by NVIDIA, brought to life by Dassault Systèmes. So I hope you got it. It's not me telling this. Jensen, you know, the founder and the CEO of NVIDIA. And I think what do we do together? We are building the foundation for industrial AI. And as he said, it enables three things: first, in research and innovations, to develop the models that simulate the causality, not only the correlation from a statistic standpoint; second, the factory of the future, which are software-defined. And why they are software-defined? Because autonomous factories continuously optimize through simulations is the reality. And third, the new way of working, which is how you can have skilled Virtual Companions, not a simple chatbot, but industry-trained experts who could help teams to design, comply, and optimize everything you do. So in 2026, now we turn this into reality with the three AI-native solutions, the new categories. The first one, we call it the Virtual Companions. They are not assistants. They are experts. They scale the knowledge. They democratize the expertise. They turn the complexity into productivities. The second one, the Generative Experiences, AI that encodes the best practices, science and compliance by default, faster innovation, lower risk, higher confidence, and third, the Virtual Twin as a Service. We don't sell the software. We deliver outcomes. That's why we are evolving our business model from seat-based licensing to value-based monetization for this new category of solutions. Because together, I think our AI-driven solutions unlock three powerful levers of value creations. The first one is expanding the adoption with the Virtual Companions, with a usage-based. Keep in mind that right now, the real limitation we face is the number of people being skilled to use our software. Now, if we have Virtual Companions being trained by design, they will take the benefit of what we deliver to them. The second one is monetizing the know-how with Generative Experiences. For 40 years, we have invested in many, many industries: aerospace, auto, now Life Sciences, High-Tech. And we have accumulated a lot of industry know-how, how to design things, how to produce them. With this knowledge, we can automate many things. And what the gain we have in front of us, it's a moonshot. It's 10 times. It's not a 20% improvement. It's not 30% improvement. It's really a radical change. Last but not least, we can sell the outcome with a Virtual Twin as a Service. Why so? Because right now, how does it work? We provide the software. You have a lot of services which are needed to implement the software on-premise, to train the people, to change the processes. Now, with AI, we can, rather than sell the tool, we can automatically generate the end result, the Virtual Twin. It's a way to reintegrate part of the value in our software. So that's what we do. And this is how we are turning AI from a promise into a concrete, sustainable value. And I think this is just at the beginning. Now, you know that was the key question a week ago. I mean, all of you, you were asking the question, which is, you know, AI is revolutionizing everything. But there are two kinds of companies: the one that compounds and the one who could be commoditized. I think Dassault Systèmes is built to compound. We are not just participating in the AI revolution. We are really shaping it for the industry. This November, at our Capital Markets Day, we will come back on this, how this vision is translating into the financial impact. But before that, now it's time for me to hand over to Rouven. Rouven, you have the floor. Be careful. Thank you, Pascal. And also, a warm welcome from my side to all of you here in the room and joining us online for our Q4 2025 earnings conference call. Before reviewing the numbers, I would like to highlight three key themes that define 2025. First, sustainable growth, our core industrial business. I want to reemphasize that our core industrial business was resilient in 2025 with strategic client wins. However, we faced a backdrop of tough comps and complex macros, specifically in the fourth quarter. We are focused on further strengthening our growth model while we are looking at improving our operational excellence. We have identified the challenges, and we will now execute to deliver, as Pascal said. Second, AI at the core. The collaboration with NVIDIA reinforces our leadership position in industrial world models, and it supports the development of next-gen AI-driven solutions for engineering and manufacturing. In 2026, our focus shifts from product launches to monetization. Third, business model evolution. The 3DEXPERIENCE platform continues to drive the transition towards cloud and subscription and our recurring revenue base. As AI adoption accelerates, the business models are evolving beyond traditional seat-based pricing towards a usage and value-based model. To better reflect this shift, we will begin to report an annual run rate, or ARR. I will talk about this in more detail later. So, as Pascal said, 2026 will be a year of execution where we will strengthen our foundation and our full-year guidance for the total revenue. So, our full-year guidance for the total revenue growth of 3%-5%, which is important to highlight, provides the room to navigate current challenges and to prepare the organization for the new area of growth. Now, with this in mind, I will transition to the numbers in more detail. In Q4, total revenue rose 1% ex-FX to EUR 1,682,000,000, with software revenue slightly up by 0.3%. We navigated a complex macro dynamic with weaknesses specifically in France and Germany, mainly in the auto sector. Plus, we also faced headwinds at Medidata and Centric in the fourth quarter. Now, we have taken the actions to address these issues, which I will discuss shortly. The recurring revenue rose 3% in Q4, with 4% subscription growth, while services were up 11%. The operating profit for the quarter was EUR 622,000,000, with a healthy operating margin of 37%. It's up 90 basis points, ex-FX, thanks to the productivity gains that we leveraged across the group. We had initiated those already entering into the year, 2025. EPS was 0.40, up 9%, ex-FX. For full year 2025, we saw total revenue of EUR 6.24 billion, along with software revenue growing at 4%. Recurring revenue grew 6%. It's creating an 82% recurring revenue base as a percent of software revenue, while subscription revenue grew 11%. We delivered good profitability in 2025 with an operating profit of EUR 1.994 billion and an operating margin of 32%, achieving 40 basis points of improvement versus last year, with an EPS of EUR 1.31, up 7%. Now, turning to the growth drivers. The 3DEXPERIENCE platform is at the core of our growth strategy and the foundation to reveal the power of AI for industry. 3DEXPERIENCE revenue grew 10% for the full year, and it's now 40% of software revenue. As expected, the fourth quarter was impacted by a strong comparison base year -over -year. And on top, we faced the weak auto sector in Europe. However, important to highlight, we signed several strategic 3DEXPERIENCE deals that have the potential to expand over the course of 2026 and 2027. This will generate future revenue and help build the momentum in ARR, which I will come back shortly. Cloud revenue at the group level grew 9% in Q4 and 8% for the full year, with 3DEXPERIENCE Cloud growing 38% and 32%, respectively. This strong growth highlights the value of the platform for clients, where the transformation is critical, as is the need to leverage AI. Now, looking at our geographies and product lines, the Americas rose 3% in Q4 with a good performance in High-Tech and transportation mobility in the Americas. Full year 2025 was up 5%. We faced year headwinds in Life Sciences, as well as in Home & Lifestyle. The core industries in the geo were strong and resilient, with 10% growth. Europe declined -5% in Q4, but it was up 2% for the full year. The weakness in the quarter was against a strong base comparison. It was also impacted by softness in France and Germany. I mentioned that before, mainly driven by the challenges in the automotive sector in these countries. Meanwhile, Southern Europe was resilient. Northern Europe gained momentum with a good performance in High-Tech. Asia was robust. We grew 6% in the quarter. It was 5% for the year. Growth was driven by transportation mobility and High-Tech. We had very good momentum in Korea, as well as strong growth in India, while Japan delivered solid growth. China had a softer quarter on a backdrop of tough comparables in the quarter. Now, to the product lines. So, Industrial Innovation was up 1% in Q4 and 6% for the full year. As noted, the quarter was impacted by the lower growth in 3DEXPERIENCE and, in particular, the challenges we faced in Europe. But overall, for the full year, we saw very positive momentum, which was led by solid traction in SIMULIA and ENOVIA and continued solid growth with CATIA. We're confident in the resilience of our core business, which is led by the cycle of 3DEXPERIENCE adoption, while preparing for the next wave of growth with AI-based Virtual Twins and companions. On the mainstream side, growth was 1% in Q4 and 2% for the full year. Growth was again driven by strong momentum of SOLIDWORKS, which was up high single-digit in the fourth quarter and in the full year. As expected, Centric was down double digits in the fourth quarter on a high comparison base. Two effects that played a role on Centric. First, we saw some renewal shifting. Also, we accelerated the move to cloud, as we explained to you previously. Now, we expect marked recovery this year in 2026 with a new management in place and a robust pipeline that's building going forward. Now, to Life Sciences. Here, the growth was lower than expected. We were down -4% in Q4, while full year was -2%, as we faced continued headwinds for Medidata, which I will cover in more details shortly. Outside of this, Medidata signed several strategic account winbacks over the course of the year. This includes the likes of Novartis, Merck, AbbVie, and Gilead. It highlights our competitive advantage, as we build strong foundations and expand our footprint with large pharma companies. Now, as we look ahead, we are convinced that the time has come to transform the biopharma industry from a document-based approach to Virtual Twin -based operations, as Pascal highlighted. This has been our vision for Life Sciences for a long term. Therefore, let's take a holistic view of our Life Sciences industry on software revenue that we are generating today. You see on this slide, this includes Medidata, as well as the 3DEXPERIENCE portfolio adopted by pharma and medtech. In order to better highlight the growth dynamics, we are differentiating two elements: the direct business, the enterprise business, as well as the indirect go-to-market model we have, where we predominantly sell to our CRO partners. Now, to the direct enterprise business. It accounts for 70% of the total revenue in this industry. It grew 3% in total in 2025. Now, within that, the Medidata enterprise business grew 1%. However, this growth was impacted by one client, Moderna, that adjusted its run rate to reflect lower study volumes. Excluding that, our Medidata enterprise business was, in fact, up 6%. Meanwhile, 3DEXPERIENCE grew 7%. Now, to the indirect business. Here, important to understand, this business is mainly focused on the biotech sector, on the very small pharma companies. So, selling and here, we are selling through CROs. This accounts for 30% of the Life Sciences industry. This is declining by -5% year-over-year. Our market saw lower study starts, which were down -7% compared to -5% revenue decline. The study starts were lower by -7%. Importantly, we continue to expand our market share also here in phase 2 and phase 3 by about 1 point in 2025. And now, also, we don't want to anticipate this too early. We did see some green shoots in Q4, as large CROs are starting to increase the number of new studies on our platform. So, what are the actions we are taking to reinvigorate growth? You see them here on the bottom of the slide. First, on the enterprise side. What we are doing is we are setting in motion dedicated account teams to focus on overall pharma transformation with our platform and leveraging AI. These teams are formed and in action across all the geos. Now, to the indirect business, the goal is to reduce our exposure to volatility in the volume business. To this end, we are evolving our pricing model and terms and conditions to monetize continued data access, which will be critical in the times of AI because this is the way the models will evolve and will help and support our pharma customers to improve the efficiency of clinical trials. Now, turning to the cash flow and balance sheet items. Let's start with the operating cash flow. We generated EUR 1,630,000,000 in operating cash flow year to date. This was up 1% compared to last year on a constant currency basis. Indeed, despite a challenging environment marked by FX headwinds and new tax regulations, we demonstrated resilience in cash generation. As previously discussed, we absorbed about EUR 40 million hit in 2025, which was driven equally by the hike in employer contributions on share-based compensation and new exceptional tax for large companies in France. Excluding this, operating cash flow grew 3%, ex-FX. In the first half of 2026, we expect the working capital to be positively impacted by the collection from large subscription deals we signed in 2025. Now, to free cash flow. It was up 2%, also excluding currency. CapEx investments were lower approximately by EUR 30 million due to lower investments in leasehold improvements versus 2024, while investments in cloud and IT infrastructure were stable. The cash conversion remains a top priority. We reached 82% for 2025 versus 84% in 2024. This is ahead of our previous estimates, mainly due to better collections. In 2026, we expect the cash conversion rate to improve, driven by cash collections and better alignment of billing to revenue. Now, to complete the picture, cash and cash equivalents totaled EUR 4,125,000,000 at the end of 2025. It compares to EUR 3,953,000,000 at the end of 2024. This increase of EUR 173,000,000 includes a negative full year currency impact of EUR 263,000,000, mainly due to the weakening of the U.S. dollar over the period. The net cash position reached EUR 1,530,000,000 at the end of Q4. Any additional information, you will find in the operating cash flow reconciliation in our presentation that we published this morning. Now, I want to transition to a new topic. Pascal already introduced it: the annual recurring revenue run rate, which we are introducing in 2026 for now. We already previewed that with you at our Capital Markets Day in June last year. It's a key metric to reflect our continued transition towards a subscription and cloud-based business model. We believe that ARR provides a consistent view of the underlying run rate and the health of our recurring revenue base, while it's also eliminating the volatility from revenue recognition. As such, the ARR is a snapshot, which reflects the 12-month recurring value derived from all active contracts at period end. It includes software subscriptions, cloud, SaaS, hosting, as well as support. It excludes future commitments. In the appendix, you will find a detailed definition of ARR on how the methodology is applied. We also are giving you three illustrative examples. Now, if you look at the numbers, growing at an average of 6% over the last two years, the ARR highlights a consistent execution in querying subscriptions and cloud and is driving the growth of our recurring business. It is also more closely tied to the invoicing and cash flows from those deals. In Q4 2025, the ARR reached almost EUR 4.5 billion, with an increase of around EUR 100 million of net ARR in the quarter. This highlights the consistent performance in signing new cloud and subscription contracts, while revenue is highly dependent on the timing of revenue recognition. In 2026, we are establishing this new metric in our reporting. The plan is to guide starting 2027. Now, during the Capital Markets Day in November, we will outline the steps in the context of our 2029 financial plan. Now, as we look ahead, the trajectory to accelerate growth is, of course, linked to the shift in our business model. Now, let me discuss very briefly the levers of ARR growth. First, the mix is driven by faster growth of subscription versus maintenance ARR. I think we see that already clearly in our numbers. It's happening. Second, the growth in 3DEXPERIENCE and cloud as AI-powered Virtual Twins and Virtual Companions boost our 3D Universes portfolio. The third, within Life Sciences, we are expanding our footprint. We are creating and preparing the next generation of growth with a new clinical trial platform powered by AI. Finally, with Centric, the ARR growth has a long runway. Now, with this, let me turn to our financial objectives for 2026. We expect total revenue and software revenue growth of 3%-5%, ex-FX, for the full year of 2026. Importantly, this guidance marks a tipping point. In 2026, the share of subscription revenue will surpass the maintenance revenue. That's also why we are providing an ARR to better reflect the growth dynamics, not yet as a guidance, but to show the momentum and the progress. The operating margin is expected to achieve 40%-80 basis points improvement, ex-FX, which takes us to the range of 32.2%-32.6% as we continue to balance investments and margin expansion, leveraging our operating productivity gains. We see the EPS soaring at 3%-6%, ex-FX, to EUR 1.30-EUR 1.34. Now, this is all based on our FX assumption and our full year average rate for the U.S. dollar to euro at 1.18 and yen to euro of 170. Now, quickly to Q1, we expect 1%-5% growth for both total revenue and software revenue. Operating margin is expected to be in the range of 29.2%-30.7%, and EPS at EUR 0.28-EUR 0.31. Finally, I would like to share some key assumptions underlying this guidance framework for 2026. So first, we expect 3DEXPERIENCE and cloud momentum to remain broadly in line with last year. It's driven by continued expansion within our installed base and ongoing market share gains. We are focused on entering new markets and accelerating the monetization of our AI portfolio. Now, from a geographical standpoint and industry standpoint, the demand in the Americas remains healthy, while Asia continues to show resilience. In Europe, we see solid pipeline development in southern and northern regions, which is partially offset by continued expected weakness in the automotive sector, mainly in Germany and France. Potentially, this is impacting the timing of decision-making within quarters. The defense sector represents a potential upside. Within our mainstream business, SOLIDWORKS continues to deliver mid to high single-digit growth in both revenues and users. For Centric, we expect a return to low teens growth supported by execution against a strong pipeline and a higher mix of cloud revenues. Now, Life Sciences is facing a transition year with actions underway to position the business for a return to growth from 2027. On the margins, we expect continued improvement driven by productivity gains from AI initiatives and operational excellence. So these initiatives are focused on increasing our flexibility and reallocating investment towards top-line growth. Now, in conclusion, 2025 and 2026, we are laying the foundations for our next phase of growth. I want you to remember three things. First, the 3DEXPERIENCE platform is at the core of our industry transformation. And it's creating a long runway of growth. On AI, we are introducing new categories of solutions. Those go beyond productivity gains. It's about creating new possibilities. And we are taking actions to scale our operations with one single goal in mind: to generate sustainable growth. Now, with this, Pascal and I look forward to taking your questions. We will start taking questions from the room. If you have any. Okay. Hello. Thank you for your time. Thank you for your presentation. Vishal Tarango from ING. I allow myself to start with asking the questions maybe with everybody's mind. Your guidance seems quite cautious for 2026, right? I can understand there are headwinds. There are some things we are in the beginning of the long runway. But what could go wrong? I mean, what do you think will be the major headwinds in the year to come? Thank you. Well, Rouven and I start. And then after you, you will add whatever you want. What's the difference between the 2026 guidance and the 2025 guidance? 2025, we were running quarter after quarter after the top-line guidance. And you have seen we have been able to deliver the EPS as initially planned. But we have to adjust in the middle of the year the top line. So I do not want this for 2026. Let's be clear. So maybe you will see this as a very conservative guidance. And I accept the point. But from a dynamic standpoint, I really want to build on the good momentum and, again, keep the company focused on what matters, which is in these AI stories. I mean, it's time for us to invest massively. It's time for us to take the positions. It's time for us to accelerate the transition to subscriptions. This is the game plan we are doing. That's the reason why, to be direct with you, I think we are relatively conservative in our guidance. Now, what could be wrong? I think we have factored many, many things already. I think you have seen in Rouven's presentations, compared to 2025, there are certain things which are moving in a good direction. Centric is moving and back to growth. It was really a headwind last year. Medidata, we are cautious, even if we have early signs of improvements. But we did not factor improvement in the guidance. On the mainstream, you see H1 last year was growing mid-single-digit. H2 last year is growing high-single-digit. We have a better perspective for 2026. On Industrial Innovation, I think we are taking some cautiousness on the auto sector, especially for Europe, which was, in a way, the bad surprise of Q4 last year. That's what's in it. I don't know if you want to add a few things, Rouven. For the top line, that's the situation. I think the other point that's really important to highlight is we are creating the room also to make investments to support our growth because it's very critical at this point in time. We are at an inflection point to accelerate growth. We are transitioning the business model. We are focused on accelerating our growth in subscriptions to build the ARR for acceleration. I think that's the upside we have. That's how we constructed the 2026 outlook for growth acceleration to come. Thank you, Derric Marcon. So one remark, Rouven. You told us in June 2025 that you will give us ARR, subscription annual recurring revenue, without removing any guidance or projection. But we no longer have the split of your recurring revenue target between subscription and support, something you gave before. So maybe you will continue in that way. And so first question, can you help us to reconcile the guidance, +3%, +5%, with the ARR growth that you project for 2026? And if you have an acceleration at the end of 2025 on ARR, why we don't see that on the revenue guidance for 2026? And the second question is about the remaining performance obligation that you gave at the end of the year. Can you help us to reconcile or bridge what is in the RPO next 12 months and ARR, please? Because the RPO was growing nicely at the end of 2024. We don't see that in the 2025 numbers. I'm wondering if we will have the same picture at the end of 2025 when we try to extrapolate the RPO numbers for 2026 and reconcile that with the guidance. Sorry, it's pure figure, Pascal. No, no, no problem. I will start, by the way. Rouven, feel free to add whatever you. So first of all, on the split between subscription license, if you go to the appendix, we are providing the details. So we did not change the way to guide the market. And we definitely do not want to hide something. Now, as you know, the annual recurring run rates cannot be fully translated into the revenue the following year because you have the revenue recognition mechanisms. And I will let Rouven elaborate. Nevertheless, if you look at the last two years, it's a good proxy to approximate the recurrent revenue anyway. Now, how to bridge with the guidance, which is your questions, because you have 6% on one hand and you have 3%-5%, right? So remember, in the recurrent part of the revenue, half is subscription, half is maintenance. The maintenance support is going at 1%, which basically means the rest, the other 50%, should at minimum grow to 11%-12% if you do the math. That's point number one. Point number two, why 3%-5%? Because if you are at 5%, it means the license are flat. If you are at 3%, it means the license are decreasing by 10%. As simple as that. This is how you will be able to reconcile the guidance we are providing with the ARR. Maybe, Rouven, you could explain a little bit more, some specificity. Yeah. We gave the 3 examples that I think are illustrating the methodology. What you really need to keep in mind, what is really the challenge of Q4 also is the tough comparison of Q4 2024, where we had very high subscription growth, in parts also because our on-premise subscription, also including some in-quarter revenue, which when you look at an ARR, it's all allocated over a 12-month period, right? We really only look at the run rate. We don't consider any revenue in point in time. It's really over time. And now, as you straight-line all of our bookings into this methodology, you will see the true growth of our business activity outside of the revenue recognition noise, so to say, that's in the numbers. So that, I think, creates a clear metric of year-over-year comparison. Now, over the last 2.5 years, it grew consistently 6%. As Pascal said, you have to understand the underlying, I think, growth drivers and momentum because the subscription is scoring 10% + percent. Also, keep in mind, we are still facing inside that the Medidata headwind. So the core business of Dassault Systèmes, 3DEXPERIENCE, is scoring much, much faster in the subscription revenue, which is important to understand. And as a Medidata business, it's expected to improve. We will see the upside of that as well in this number. One other comparison I would like to share with you is the recurring revenue growth was 6% over the last two years. And it's very consistent with the ARR. Now, we're talking about an ARR of EUR 4.5 billion, scoring at 6% with a potential to accelerate because of the mix effect as well as subscription is getting to the threshold of 50% +. Then you have a base effect where that growth is starting really to become meaningful. Maybe a last point. You mentioned the remaining performance obligation. I think the coverage that we have in our visibility for the next 12 months subscription revenue growth, that is very comparable to 2020 when we enter 2025 as it is in 2026. We are forecasting 8%-10% subscription revenue growth, for which we have good visibility to achieve that. Now, maybe I should add one additional topic, which is an interesting one. If you look at the performance of Q4, 1% growth, flat for the software. If you look at the ARR, it's EUR 105 million incremental over EUR 1.5 billion software revenue, which is 7%. So again, I'm insisting on this because we are really transitioning to the subscriptions and to the cloud. Now, if you look at the way many of our peers, they did it, they were decreasing for many years. We are not. We are slowing down the growth, for sure. But we are really transitioning. And that's the reason why this metric is extremely important for you because it's a way to see the momentum we are building. It's a way to see the ramp-up we are creating with those long-term contracts. This is what is behind. I think they had another question related to, no, Derric, you had the last question. Just on the remaining performance obligation. I think the next 12 months figure for remaining performance obligation last year, so at the end of 2024, if I remember well, was growing in the low 20s, so 23% or 24%. Why we don't see that appearing in the numbers of 2025? And when we will look at the figure at the end of 2025, how can we extrapolate that number and bridge that with your guidance for 2026? Well, Derek, we are not guiding in remaining performance obligation because the way the revenue recognition is, we're not fully ratable as it relates to that. The remaining performance obligation is a value aggregated for the next 12 months of bookings value, not revenue. So it's always depending on the timing of renewals, right? In a year where there are upcoming larger renewals, it is less because then it will be back to the growth after the renewal. So there can be time-to-time variances. I think what's important to understand for you is that the visibility we have into subscription growth is the same in 2026 as it was entering 2025. At the same time, you missed the guidance in terms of subscription revenue, Rouven. Sorry. At the same time, if I compare the guidance for subscription in 2025 at the beginning of the year and the end results. I understand. I understand your point. You never have 100% coverage, of course, right? The coverage is typically around 84% for subscription. So depending on the timing of signing, this can impact at the end of your achievement. And this is what happened. Morning. It's Laurent Delforge, Kepler Cheuvreux. I also have three questions. The first is on the 3D universe and your plan for the next two or three years. I was wondering if you were planning to invest mostly organically adding staff or if you were considering partnerships or even M&A, so the way you will build it. My second question is on Centric. You had a very strong decline in the fourth quarter. What can you give us to make us comfortable with the fact that you will renew with double-digit growth and probably, hopefully, in the first part of the year, maybe the pipeline or whatever? And my last question, even if I don't want to preempt the Capital Markets Day, from 2026 to 2027, what could be on top of maybe Medidata, the additional growth? Is it AI-related or any other things? Thank you. Start with the first. You start? I can start with Centric. Yeah? Yeah. So you're right. Q4, we faced the decline. The visibility we have in 2026 is the pipeline is building. The visibility for the first quarter is already there. What is also important to highlight is not only the SaaS transition but also the diversification. So we are going really from the apparels, shoes, the traditional business, the apparels, the consumer-centered industries, also to food and beverage and retail. So we are really expanding our scope. And the new leadership team is embracing that as well as the opportunity to expand really enterprise-wide from the front end also to the back end with 3DEXPERIENCE. So we have multiple growth levels on Centric that we are building and that are contributing to the business going forward. And that gives us confidence. And the last point, I would say, when you look at it geographically for Centric, the Centric business has a strong momentum in Europe. Asia is building. And in Americas, we see a strong reinforcement. And that is, I think, another very important element that we really see that business on a global basis diversified. And that has strengthened as we enter in 2026. We continue to invest. You remember, last year, we made the Quyntess acquisition. We are now through the full integration of that. And this really expands the domain and the platform for Centric to expand within the current customer base and brands, which is getting another potential points of growth. And maybe above all of this, Laurent, the problem came from the previous management who did an acceleration of the revenue. And this is clean. I mean, now it's behind us. Coming back to 3DEXPERIENCE, I think the way you should look at it, it's not an extension of the current portfolio. I mean, it's really a redefinition of our offerings the same way we did when we came with 3DEXPERIENCE almost now 15 years ago. So why I'm saying this, it's because it's requesting a new ecosystem. So in a way, you have seen new partnership. And NVIDIA is a good example of this. You need accelerated computing capabilities. At the same time, NVIDIA, they need to I mean, computing without knowledge is blind. So we are the one providing the knowledge with our Industry World Model. So the combination is extremely meaningful. We will continue to partner with basically people leading this game. The real question, I think, is the M&A behind your questions. Yeah. If we look at the landscape, I think it's not a bad time to consider M&A. Medidata reinforcement is behind us. The cash flow is relatively significant even if you are challenging a little bit the cash conversion. But at the end, it's a lot of money every year we are generating. The exchange rate is helping us to consider certain acquisition in the U.S. And the software landscape is under pressure. So I remember having exactly the same question two years ago. And I was telling you, "It's not the right time." I think now it's the time. I will not say more. But at least you understand the way we think. And the last part, which is related to the 2026, 2027 ARR, what are the growth drivers, I think? Maybe, Rouven, you just started to mention it in your presentation. Yes. Yes. Yes. And I think 2026, we have the guidance, Laurent, right? So let's look at 2027 and beyond. What are the potential upsides we have and how we can accelerate? First of all, keep in mind our large, large client base and the long runway we have to penetrate that client base with 3DEXPERIENCE, cloud, and our AI portfolio. That's really the strongest engine we have and the acceleration potential. Now, this together with the transition to the subscription and cloud, which creates the recurring revenue, building on top of that and creating the base effect of the faster-growing subscription growth. That's, to me, the first thing that is important. You see, we are focused on that to build this, to reach this inflection point where we will see the acceleration. You're right to mention Life Sciences. You see that. We are taking the investments. We have a lot of good things going already. You see, the enterprise business is growing when you exclude some special effects. So there is a lot of good things that are happening. And we will continue to double down to strengthen them. And as it relates to the volume business, once we retain that exposure to churn, we will also see the benefits here. The expansion on the 3DEXPERIENCE portfolio for the industry, Pascal highlighted that as a massive opportunity. Centric, we discussed I outlined this before, is another growth driver that is material. And then there is the last point. You mentioned it, which is potential M&A, which is not factored into this model, creating an upside that we have. So that's the situation as we are transitioning from 2026 to 2027. And this is what I will focus to explain and outline in the Capital Markets Day in November this year, where it's about translating that into an ARR model to give you confidence and understanding of the path ahead. We take one more question in the room, and then we'll go online. Yes. Thank you. Gregory Ramirez, GR20 Research. I have one question on the healthcare business for 2026. Trying to do the math, it seems to imply something low- to mid-single-digit decline. What is the implied impact of the Moderna contract? Does that mean that if the impact for 2025 is just on one quarter, that will have still some headwinds for the first three quarters of 2026? And regarding the enterprise business excluding Moderna, does that mean that it will be growing? So when we exclude the Moderna effect, that means that maybe in Q4 2026, this will drive the growth for 2027. Yeah. Maybe first, I want to clarify that we're not expecting decline in the Life Sciences business overall in 2026. We're expecting flat in 2026. With everything we do. With everything we do in Life Sciences. From an industry standpoint, we expect to be flat. As it relates to the impact of Moderna, this is a 2025 impact. It was not only related to Q4, but it's really for 2025. But the aggregate of this impact is now behind us. Moderna reduced their contract by more than half to reflect their new business-level activity. We're still the prime partner for Moderna. We have not given that or losing that to anybody else. It's our client. But their business realities have changed. And we have adjusted to that, which had a big impact on the performance. So now, with that behind, as we look at the enterprise business going forward, I gave you some names and large enterprise clients that we have signed where we are expanding our footprint. And beyond that, we will do even more. By the way, in March, we have Medidata Next where we have many of those clients coming with us on stage to explain what they do with us, to expand across our portfolio, to transform what clinical development is today. That will translate. We're confident about that. That will translate into growth in the enterprise segment going forward. Where we have less control is on the part of the volume business with the CROs. This is where we are, to some extent, exposed to funding environment on the biotech sector, which can be volatile. We know it has not been great over the last two years. But also there, we see some green shoots coming. If that materializes, it will stabilize the business. Maybe one additional thing. So if you look at the enterprise segment, and Rouven showed the numbers to you, the structural growth of Medidata, excluding Moderna, is 6%. And the growth of the rest of what we do is 7%. So we have two legs, if you want, which are solid. And we do not see a reason why it will not continue to be the case in 2026, right? The part where we are extremely cautious because we have been burned last year with this. Not only last year. We were betting recovery of the CRO to be back to at least being flat. We landed at -5%. That's what is not a factor in the guidance. We do not take any improvement in the guidance of the CRO business, okay, despite the fact that some of them, they are publishing better results in Q4 compared to Q4 2024. Thank you. Thank you, Ilona. Take questions online, please. Please, could you start the online Q&A? Yes, of course, dear participants. As a reminder, if you wish to ask a question, please press star one one on your telephone keypad and wait for a name to be announced. To withdraw a question, please press star one and one again. To ensure everyone has the opportunity to ask a question today, please limit yourself just to two questions at a time. Thank you so much. Now, we're going to take our first question. It comes to Mohammed Moawalla from Goldman Sachs. Your line is open. Please ask your question. Yes. Great. Thank you. Morning, Pascal. Morning, Rouven. My first question was just to sort of understand the 2026 outlook a bit better. Can you help us understand at both the low end and the high end of the range, kind of the assumptions, particularly around what you're incorporating for some of the larger deals as well as the kind of mega deals, what's in there and what isn't? And then, again, coming back to sort of bridging the kind of ARR to the revenue growth guidance you've given, you sort of mentioned that we're kind of getting close to parity between subscription and maintenance. But obviously, in there is the Medidata and Life Sciences recurring revenue. So would you give us a better feel for what the underlying subscription growth rates in kind of the core of the business ex Medidata is given? We're doing a transition both in the enterprise business, but also as well as Works and Centric, just to sort of better understand the moving dynamics of growth. And then lastly, you sort of touched a bit on more consumption outcome-based revenue as you drive kind of the AI applications or use cases. How does this sort of change kind of the visibility going forward? Because I think because of conventional business, there's going to be more variability. But when does this become kind of more meaningful to your mid-term growth rate? And is that going to be and how is that going to be captured? Thank you. Okay. Host, thank you for your questions. I start with the first one on how to position the guidance of 3%-5% revenue growth. As I said at the beginning of my prepared remarks, I just want to reemphasize this point, that this guidance provides us room to navigate. It's de-risked. It's de-risked for mega deals. We do not include any mega deals into that. Of course, we always have large and chunky deals that are important to happen, but not mega transactions. I think what we need to do to overperform this guidance is I mentioned that there are certain areas that are introducing upside into what we do. We have not included significant growth in the defense sector, for example, simply because we have not yet seen this sector to contribute incrementally strong to our demand. But there are opportunities, of course, in this market that we are very focused to tackle and to get involved and to take benefit of it. We are building a very powerful AI portfolio, transforming our industries and building that to expand what we do with our current clients. And Pascal gave the examples and discussed that this has, of course, huge potential for us to accelerate the growth in 3DEXPERIENCE and AI. For now, as you see in the guidance, we have taken a more prudent approach to and this was what I shared with you as well. We consider the momentum in 3DEXPERIENCE growth to be consistent with 2025 and 2026. So of course, now, as we enter in 2026, we are a significant step further in making these use cases and scaling them and replicating some. There is, from that perspective, clearly opportunity for us to improve and growth and accelerate from there. Now, we have to keep in mind that we have to offset also some potential softness in the auto sector. Now, there are two sides to look at, two ways to look at that, while there is maybe softness in the fourth quarter. But still, it represents significant opportunities for us also in 2026 because, I want to remind you, Mo, and everyone here, that we have signed significant deals in this sector over the last two years that are building the foundation to bring these clients onto 3DEXPERIENCE, now with the potential to expand. And we are expanding with them in AI. We are changing the game with that and creating significant opportunities that we couldn't do years before because we simply weren't in this position to do it. That is all in front of us. Now, in this context, I think we wanted to position a guidance from 2026 that is consistent with 2025. We ended at 4%. The midpoint of this guidance is at 4% with the potential to do better than that, but also to have the room to make the right choices for the mid to long term to accelerate growth. On the low side, what can happen on the low side more? There are macro factors that can always impact us and make things even more difficult and the timing of decisions more difficult. That can be a factor that we took into account on our guidance to be at the low end of 3%. But that's not what we are targeting. So going to the ARR, you're asking for a lot of additional information, Mo. And for sure, I will be prepared to disclose some of that at the Capital Markets Day. So please bear with me that I will be a little bit more high-level. But I will give you the directions on where the ARR, what are the elements of the ARR and the different growth dynamics inside the ARR. You're right. When you call out Medidata as part of the ARR, of course, that has been a headwind on the 6%, very clear, on our subscription ARR. I mentioned subscription ARR is growing more than 10%. Now, when you exclude Medidata on core industrials, subscription ARR is growing mid- to high-teens. It's a strong, resilient growth driver for a long time. And as I said, we have the potential to further accelerate that as more and more of our business transitions to cloud and 3DEXPERIENCE. Now, you know the size of Medidata. Medidata, in total, is about a EUR 1 billion business. Now, in the way we've defined the ARR, we have some business in Medidata as it relates to the volume business where we have churn effect. That's not renewable. It's not included in our ARR. So we have not factored in the 100% of Medidata subscription business because not everything is renewable. But nevertheless, it's a very large number inside the subscription ARR that has been facing a lot of headwind. Once we are removing this headwind, and you see we are already doing it for the enterprise part now, the CRO part is still what is the challenge. But as we are removing that and increase and essentially generate a net increase in ARR quarter -over -quarter for Medidata, that will then contribute to also increase the subscription ARR from what today is +10% to the next level. And I would say to go from +10% to gradually up point by point to reach 15%+ in the subscription ARR in aggregate. But again, I will be prepared to outline those components at the Capital Markets Day in more detail. Here today, for me, is to introduce this concept to you, explain the growth drivers, and with that, build the foundation on the levers that we are focusing on for growth acceleration to come. Now, the last part of the question, Mo, is related to the new business model for the new AI solutions. If you look at the story of Dassault Systèmes, we changed almost everything except one, the business model, because it has always been an equation of the number of users, the number of seats. Now, what's the problem? With the AI, you have virtual users now. We call it Virtual Companion. Some of our peers, they call it agent. So the model cannot be anymore the same. That's point number one. Point number two, I think the software are really selling the capabilities. Now, we have an ability, by combining the capabilities with the knowledge, to sell the end result, whatever the Virtual Twins of the improvement. So that's the reason why we are coming with new units. And let me explain to you what the units are about. The first one, we have the unit of Works. When a Virtual Companion is working, in a way, he is working like a human. And the same way you have a cost for a human, you should have a cost for the companion. So we are not inventing anything on this one because this is what ChatGPT, Anthropic, this is exactly what they do. It's a fraction of the cost of the people. More importantly, when we have a Virtual Companion, it could be a mechanical engineer. This one could be a mechanical engineer and an electrical engineer. It could be also, in addition, a specialist to do the simulation. So we have what we call the unit of knowledge. And the more knowledge the companion, they have, the more we will price for this because it's a way to capture the value of what we bring, right? If you have to hire someone right now who is graduated from a best mechanical school, graduated from the best IT or computing school, it will be difficult. With the companion, we have an ability to enable this. Last but not least, there is a third unit. We call it the unit of knowledge. And again, I insist on this. For four decades, we have accumulated a lot of industry knowledge. We know how to design a car. We know how to produce it. We know how to certify them. We know how to scale the production systems. And this is true in every industry we serve. Now, what can we do with AI? We can automatize those processes. And I can tell you, this is a significant value we are creating. In many cases, it's a 10X. It's really a moonshot. So that's the reason why we need also to have another unit if you want to price in order to capture this value. And last but not least, I make this in my comments, but the entire industry is pushing to do software as a service. I pretend we should do service as a software because you have so much money being spent in services just to make the technology enable that there is a chance now to use artificial intelligence to produce automatically the end result. And that's what the Virtual Twin as a Service is about, is how we can mix the capabilities with basically what used to be human-driven, and now it could be a Virtual Companion, in order to deliver the end results. That's the new lever we have in our hands to create additional value. It's tangible. It's concrete. It's not science fiction. It's fiction with science. And it works because it's a way. I mean, I have enough experiences the last year with many engagements to tell you, we know how to objectivize the value discussion with many of our customers. Now, this will come on top of what we do because, again, I repeat myself, if you have a Virtual Companion, you still need CATIA. The Virtual Companion is the one using CATIA. He's not substituting CATIA. You still need to use CATIA, SIMULIA, DELMIA in order to produce the end result. So that's an additional lever we are creating. And if you remember what I say, with the companion, we are accelerating the adoption. It's taking too much time to train the people. It's taking too much time for many of our customers to hire people. Now, we have a way. We master in a much better way the cycle of adoption. We have a better way to monetize the knowledge and the know-how we have put in our software because usually, people, they are comparing the capabilities, but they forget that we have put a lot of industry knowledge in our software. And again, we have the way to monetize the end result of what we do. That's the entire purpose. Now, given the visibility, it's a little bit early to speak about this. The only thing I can tell you is just last year, in six months, because we released those products, the first initial drop was mid-year last year, we have been able to basically create a EUR 50 million backlog, right? That's what we have been able to do. And this is growing extremely fast. But again, same story. We will come back with more insight at the Capital Markets Day in November. We can take the next question online, please. Thank you so much. Now, we're going to take our next question. The question comes to the line of Adam Wood from Morgan Stanley. Your line is open. Please ask the question. Hi. Good morning, Pascal. Good morning, Rouven. I've got two, please. And just first of all, thinking about more traditional kind of metrics of visibility in the business, could you maybe help us with pipeline coverage in a sense today? And I guess particularly given the comments around automotive and investor fears around that, there have been any notable shifts in terms of where the pipeline is by industry that might reassure. And then secondly, I think we've seen in software a lot where there's been deceleration in revenue growth, and particularly where companies have a lot of things changing and opportunities. You've obviously mentioned AI, the subscription transition for you. Can you just help us understand the balance that you're trying to strike between making sure you're reinvesting at the right level in the company to drive that revenue acceleration and maybe protecting margins in the short term? Are you comfortable you're getting that kind of split right? Thank you. Thank you, Adam. I will start with the first part of the question. Rouven, you will take the second one. Related to the pipeline, the coverage right now is 2.5 times. If I look at we have the pipeline to cover 2.5 times the sales plan for the full year, which is good. The idea is to create half a point more during the year because we are creating also the pipeline over the time. As a starting point, it's a good ratio. What's the difference compared to last year is the mix. Last year, we were relatively overweight with a lot of opportunity in the auto sector. If I remember well, the pipeline in the auto sector was around 35%-37% last year, right, Rouven? Yep. This year, we are below 30, which is, I think, better given what is happening. I mean, the volatility we have, especially in Europe, on this topic. Where we see a share which is increasing is in the Aerospace & Defense, which is, I think, close to 17%-18% for this year. It was a little bit, let's say, last year, it was around 12%, if I remember well. What's the difference? We still have the backlog because, as you may know, these industries, they still have a lot of planes to produce. And they still struggle to deliver the backlogs. And we have a lot of demands coming from the manufacturing part. You remember the large deal we signed with Lockheed Martin a year ago. It was on this topic. But also, we have new segments. Defense is one of them. But also, you have the New Space. New Space is, in this industry, what the EV was, let's say, 10 years ago for the car industry, people developing drones, low-orbit satellites, the new launcher, I mean, the New Space station as well. And they are the one, basically, equipping themselves very rapidly. And they are raising and growing fast their engineering departments. So this is also something important. We'll see also more demands in the High-Tech sector. It's something I did not mention, but we spoke a lot about NVIDIA as a partner. But NVIDIA is also a customer, right? And for the one who have I mean, look at what Jensen said last week on stage. NVIDIA is using our technology. In fact, we have built the Virtual Twin of their future AI factories. And if you look at this entire sector, this is where the money flows right now. I mean, every morning, you open the newspaper, you see all the hyperscalers, all nations committing themselves to invest billions to create these AI factories. It's very complex objects. I mean, you cannot imagine the complexity of it. It's much more complex than a nuclear plant. And by the way, as I command, do not make the confusion between the data center and the AI factories. The data center is, to make an analogy, the warehouse. This is where you store the data. The AI factory, it's a real factory. This is how you transform the data into insight, knowledge, value added. So the complexity and the requirements, it's I mean, there is nothing comparable to what we know currently with the data center. So where I want to go, they need a Virtual Twin. And that's what we do. We are building the Virtual Twin of the AI factories of the future. This is a tremendous opportunity where we are expanding a lot. To come back to the initial questions, that's the reason why the share also in the High-Tech sector is increasing significantly in the pipeline. We are around 15%. So long story, a short one, good coverage, mixed difference. I think the mix is probably more resilient compared to last year, less exposure to the auto sector, much more distribute. The rest, industrial equipments, the consumer goods and packaged goods, is almost the same than what we had last year. Okay. Then to your second question on the investment policy, how is this reflected? Really, we are looking at the different parts of the organization, starting with R&D. Here, the focus is on allocating our resources to accelerate the AI portfolio and also help to prepare the go-to-market with our brands because the R&D and the go-to-market, as you can imagine, as we are now bringing these next-generation applications to our clients, where R&D, brands, and go-to-market are extremely connected with our industries. So when I look at the opportunities on where we are investing on the go-to-market side, Pascal mentioned High-Tech, data centers. That's a growth opportunity. But also related to that, the energy sector and the energy transition, which is a massive opportunity for us. We have had already in 2025 some great wins here. But we know the energy shortage and the pressure on the energy market will further open opportunities for us to virtualize and to improve and help this industry. Aero and space is a very resilient industry for us where we are doubling down our focus on the go-to-market as well. And of course, industrial equipment, we have seen very resilient results in SOLIDWORKS. And we expect that momentum to be also in 2026. And last but not least, I mentioned that before, Adam, but just to complete the list here, the auto sector remains a critical industry for us with a lot of opportunities. And we are focusing also our go-to-market on that because we have a large footprint and the opportunity to expand. Now, we talked about the Life Sciences go-to-market where we are investing with dedicated integrated account teams. We are really reallocating our resources to address that industry more holistically and open up new pockets of growth and budgets that we didn't systematically address in the past, which we have an opportunity to do, and we will. And then the last part, I think Pascal mentioned that also before, we are focusing on scaling the indirect channel where just through the value network, the suppliers serving the OEMs, we have a lot of opportunity in this market as well where we are enabling our partners with our new AI solutions to transform these suppliers. Now, on G&A, we have initiated a large transformation project here as well to leverage AI and cloud in our back office to streamline the operation. We will take one more question online. Thank you. Now, we're going to take another question. It comes to the line of Michael Briest from UBS. Your line is open. Please ask the question. Yes. Good morning. Thanks for squeezing me in. Just thinking about the CMD in November, are you currently reiterating your 2029 financial ambition? Should we expect that to be updated at that event? Because I'm looking at subscription, and you're expecting it to be 55% of software by then. And it was 40% last year, maybe 42% this year. That looks like a steep ramp. And then I appreciate the comments on M&A opportunities. But given the share price, what is your view? Is there any shift in the view on potential buybacks given the valuation today of Dassault? Thank you. All right. Thank you, Michael. On the Capital Markets Day, regarding the 2029 financial ambition, there's two things. First, the introduction of the ARR will allow to have a more focused discussion on the growth levers and our path to accelerate top-line growth, which is really around subscription, recurring, cloud. And now, with AI, we are in a different point than compared to last year where we have very concrete monetization examples on how they will build on top of our existing model. The second point I would reemphasize is that our financial commitment is on the EPS. And that also has a level of the operational efficiency that we are implementing. That's why we reemphasize that today, that we are taking the actions to improve the EPS and margin, which then, of course, will have a stronger effect as the top-line comes back. We will discuss that at the Capital Markets Day and how we are going to bridge to reach our financial ambition by 2029 as we outlined last year. The second part of your question, Michael, is related to the capital allocation. So again, there is no—I mean, I have nothing against the share buyback. But if you step back a little bit, let's assume I'm spending EUR 500 million to do this. The levers on the EPS will be relatively limited. And the same EUR 500 million invested to acquire new AI startup will deliver a better lever. So that's the reason why right now, I'm really directing the capital much more to the external growth than to the share buyback. The share buyback for us, if you remember our policies, is only to offset the dilutions coming from the LTI. That's what we do. But again, I have no dogmatic positions. I'm not against. I'm with you on the value creations. But I think we have a better road to create value by investing on the new AI domain than rather than to buy the shares. I think it's time to conclude. Again, thank you, all of you, for your engagement today. But I want to leave you with one clear message. I think Dassault Systèmes is really undergoing a profound and deep transformation of its business model. It's obvious that we are transitioning decisively to subscription-driven future. And it's not only coming from the acceleration of the cloud platform strategy. But above all of this is the evolution of our offerings. AI, it's at the core of the platform. This will enable our customers to create, simulate, operate Virtual Twin at scale, to bring the virtual in the real world together and to manage for the entire industry. And I think this transformation is not incremental. It's really a fundamental transformation. We do all of this with a certain resilience. Whatever you say, we continue to grow, maybe at a moderate pace. Yes, I accept this. We sustain the momentum when, if you look at in our industry, many of our peers, they struggled when they did this transition. They did the transition before the AI came. I think I hope you get a better sense of what we do, how we are executing, how we are innovating, and how we are advancing to deliver the next phase of growth. I think Rouven and the investor relations team are looking forward to seeing you in the coming weeks because they will do roadshows and see you no later than next quarter for me. Thank you.