The current development within artificial intelligence marks the transition from a phase of pure scaling to a phase of functional differentiation. Previous approaches, in which large language models (LLMs) run through complex thought processes autonomously step by step, lead to high quality results, but cause massive economic, compliance and energy overheads due to their enormous size.

In practice, the use of large language models often fails due to the efficient processing of extensive contexts. The sheer volume of data does not automatically lead to higher decision quality; the opposite is often the case. The correct selection of data as a foundation for reliable decisions therefore represents a strategic challenge with regard to the use of artificial intelligence in the context of language models.