To keep it with Steve Ballmer very short, I was witnessing him live on stage. He said to the auditorium, “70% of you will be redundant in 18 months’ time.” He laughed about this the way our Steve is laughing. There is really a revolution ongoing, and I am very happy that we are prepared since three years on this. We are applying this massively. We acquired the best brains available from Google for the topic AI and GenAI, and I split this in two pieces. This year on the internal roadmap, the first priority is our engineering. 100% of the code from the engineers is AI-supported. There is no way for the engineers around this. This increased our lead time already by 20%, and I expect a significant uplift on those numbers. The second focus area is operations. Why should a human being not be replaced by a programmed AI agent if the data is there, if computer vision is there and video is there? There is no reason for it. We are already on 50% of our content produced with our own agents, which are working on computer vision to extract that information, tens of thousands of more data points per match, depending on the sport. These are the two key focus areas. I would love to see much speedier deployment of agents in the financial sector and in the legal sector. I think that is a big, big benefit. Statistical KPIs are something which is from yesterday. You are going to need to adapt to the market. You need to understand what is in the market and how that compares to the internal KPIs and numbers. These are areas which are not the top priority, but we are working very hard on this. Engineering and operations are the top internally. Externally, the real exciting thing is what we can do with all the data points and the deep data which we have. For this, we started now the foundation model for basketball. What it is, we have seen this foundation model because we feed the tracking data into this, and that is billions of data points. When LeBron is, for example, doing a no-look pass, we see if he is looking up, down, left, right. The human being cannot see this, but based on this and based on the position of the other players, we predict now the next pixel and the next pixel, and from this we can predict seven seconds with a high probability from where the points are scored. Imagine how cool this thing is for coaching applications, but imagine how cool it is for market makers to have that edge to predict what is the potential next move and underline it with liquidity. This is exciting stuff, and that is now possible with AI. I could speak hours about this, but these are the main sectors where we are looking into.