Thank you for your query.
Nokia encourages all its diverse functions to propose AI use cases based on their requirements. Deciding whether to utilize pre-trained AI models (such as those available on Hugging Face) or develop our own AI solutions depends on multiple factors, including telecom-specific use cases, resources, expertise, and the level of customization needed. We have a dedicated team that analyses and prioritizes use case development based on operational efficiency that can be achieved through its implementation. In some of the use cases where sensitive data (Both Nokia’s and Customer's) is involved, Nokia plans to create their own LLM. But if creating an LLM is not feasible, Nokia goes for a Pre-trained AI model available in the market but only if it can be hosted locally thereby ensuring no data leakage to the outside world. Hope this clarifies.

14 October, 2024 14:08