Behind the anthropomorphic vocabulary are probabilities, vectors, and learned weighted averages A scientist who has spent decades studying nonlinear systems, modelling, statistics, or linguistics can open a modern paper on large language models and feel that an entirely new branch of mathematics has appeared. Query, key, value, attention, heads, embeddings, transformers. The impression is misleading. … Read more
The rapid development of Artificial Intelligence is making something increasingly visible that we often tend to overlook: the digital ecosystem has a very tangible physical foundation. AI models, cloud services and public digital applications run inside buildings filled with processors, storage systems, power supplies, networks, cooling loops and backup power systems. Data centers consume electricity … Read more
Why their value for the IT market lies in reuse, open standards and avoiding vendor lock-in In the IT sector, we know well that a successful prototype is not necessarily a sustainable information system. It may work perfectly in a controlled environment, yet fail when it has to integrate with existing infrastructure, be maintained for … Read more
The future of AI is not something that will simply happen to us. It is the result of choices about who designs technology, who owns it, who controls it, and to whom they are accountable. One of the most dangerous ways we talk about Artificial Intelligence today is as if it were a natural phenomenon. … Read more
The trap of the US-China AI race The American debate on artificial intelligence is increasingly shaped by one simple but dangerous sentence: if we slow down, China will overtake us. This sentence has become a political shortcut. It allows major technology companies and their allies to describe regulation, safety testing and public accountability as threats … Read more
A breakthrough, not a final destination Large language models have already changed how people write, code, search, translate, summarize, teach and organize knowledge. Their success rests on a powerful empirical insight: when models, data and compute grow together, new capabilities appear. This is the core intuition behind the scaling hypothesis, and it explains much of … Read more
Europe does not suffer from a lack of open source. It suffers from a lack of large-scale adoption. Across the European Union, the policy direction is already clear. Interoperability, reuse and cross-border public service improvement are now central to Europe’s digital agenda. Open source has also been increasingly linked to reduced dependency, stronger digital autonomy … Read more
Open standards for documented linguistic knowledge Language corpora have become a foundational infrastructure for linguistics, natural language processing, and contemporary artificial intelligence. The term corpus does not merely denote a collection of texts but implies deliberate selection, structuring, and documentation according to explicit design criteria. Within this context, the Text Encoding Initiative Guidelines provide a … Read more
Apertus AI is one of the most transparent and technically mature efforts to build a fully open-source large language model. Developed in Switzerland and released together with its source code, training documentation and model weights, it offers an unprecedented level of reproducibility and independence from closed ecosystems. This makes it ideal for researchers, public-sector institutions … Read more
Government emphasises open standards and urges use of liberating alternatives Sweden’s public administrations should use open IT standards and software “that frees the state administration from dependence on single technical solutions”, the government writes in its note on ICT, part of its 2016 budget plan. “For eGovernment solutions, IT standards will play an increasing role … Read more
