CASE STUDY
AI is different from previous technologies because, particularly when we are talking about Generative AI, a lot of what we are doing involves modelling your intellectual property and the way you operate as a business into that AI.
Analysts are naturally cynical of ' black-box ' AI because you can ' t secure what you can ' t audit. How do open source and Red Hat’ s open hybrid cloud provide the foundational transparency that proprietary platforms lack?
Our platforms are open source. When we talk about Generative AI specifically and the large language models you might rely on, there are important questions to consider. Do you own the model? Do you know what is inside the model? Can you replace that model?
That ability to replace models is particularly important because, as everybody has noticed, the technology is still evolving very rapidly. Every three, six or nine months there is another advance in the technology. Sticking to one fixed model could potentially hold you back. Being tied to a single model could create another problem.
Everything we are doing is therefore about providing the choice to select your models and decide where you want to run them. When we talk about hybrid cloud, there is a need to be able to run things in the public cloud. There is also a need to be able to run them in a sovereign cloud, which is similar to the public cloud but operates within your sovereignty, giving you greater control and the jurisdiction of the laws that support it. You may also need to run them on-premises.
Having that flexibility means being able to decide what the best choice is at any particular time for your business and your own business processes. That is what we are focusing on, while helping to remove the friction and level the playing field for everybody.
AI is one of those technologies that can change things significantly. If you lower the barriers to access and make it easier for everybody to use and run the technology, you level the playing field so everybody can compete on the same terms. It is no longer about simply having access to AI. It is about being able to use it in the most advantageous way for your business.
The sovereignty framework relies on four pillars: Data, Technology, Operations and Assurance. How does this architecture practically guarantee absolute control over data and models, protecting enterprises from foreign infrastructure dependencies or partner disruptions?
This is not just a Red Hat model. It reflects the way the industry is thinking about sovereignty. If we start with data, everybody was talking about
16
INTELLIGENT CIO EUROPE www. intelligentcio. com