Companies are moving to open source AI for independence

The era of open-source models in artificial intelligence has begun, as reported by Zamin.uz.
Klem Delangue, head of Hugging Face—one of the largest AI platforms in the world—stated today that business representatives no longer wish to remain dependent on closed systems of major tech corporations. According to him, modern companies prefer shifting to open-source models to preserve and develop their technological independence.
These details were provided by TechCrunch. The Hugging Face platform currently serves as a specialized hub for AI developers.
Here, specialists can exchange ready-made models and datasets. The fact that nearly half of the companies listed in the Fortune 500—ranking the world’s most influential firms—use this platform’s services indicates a significant shift in the balance of power within the AI market.
Economic efficiency and control are among the primary reasons driving companies toward open-source solutions. Typically, business projects begin by using ready-made external services.
However, as projects scale and user numbers grow, costs for such subscription-based services can rise uncontrollably. Sharp increases in expenses compel companies to seek open-source models that can operate smoothly within their own infrastructure.
This approach not only saves money but also ensures data security and allows systems to be precisely tailored to specific needs. Furthermore, Klem Delangue expressed concern that the AI market risks becoming concentrated in the hands of just a few large corporations.
In his view, if the sector develops only around closed systems, it could lead to the stagnation of innovation and the decline of healthy competition. The suspension of some major projects also signals instability in the market.
Open-source models, on the other hand, return full control over intellectual property to companies. This trend is also highly relevant for Uzbekistan’s rapidly developing information technology sector.
Local startups and government agencies should adopt open models to avoid reliance on costly foreign services. Adapting such models to the Uzbek language and local conditions represents the most favorable path toward ensuring the country’s digital sovereignty.





