So, I don’t think data competence can be taught. It can only be learned. And the learning begins as early as possible. So, we’re very big on so-called edge AI or open-source AI. In fact, for public sector use, we advised against the processing of personal data with cloud-hosted AI services. If you want to process personal data like in an email, I fine-tune the model on my own laptop. And the inference, everything is done in airplane mode. And so that security-wise, it doesn’t go anywhere, of course. But also, competence-wise, it means that people don’t see fine-tuning and alignment as random jargons. That’s something that they just do every day. And I think this is important. People are going to have AI competence in next-generation workplace.