With great powers from the US to China vying to dominate sovereign AI, technology is no longer a tool, but a mere geopolitical asset. Countries are remapping AI on behalf of the nation, sparking essential issues regarding censorship, disinformation, and ideological control.
The US executive order, “Removing Barriers to American Leadership in Artificial Intelligence,” is a proclamation of Washington’s fixation on becoming the global leader in AI data localization. Former orders, such as the National AI Initiative of 2020, had reaffirmed that AI must stand for “fundamental American values,” injecting political agendas into technological innovation.
The quest for dominance has birthed sovereign AI, when large language models (LLMs) are trained on local data sets and cultural nuances, to produce AI that speaks from national perspectives.
As NVIDIA CEO Jensen Huang told Bloomberg, “Europeans want an alternative to US-centric AI models,” referencing growing frustration with American AI nationalism dominance.
Yet this trend is not risk-free.
China’s DeepSeek LLM, for example, won’t answer questions on politically sensitive topics like Taiwan or Tiananmen Square. “It’s not a bug but a trained algorithmic pattern,” critics complain—one that shows state controlled AI censorship.
Controversy in America also trailed Musk’s Grok AI, which gave anti-Semitic responses to prejudiced questions and the lack of ethical filters. Later, Musk’s company claimed that Grok had been retrained to “respond to political incorrectness with overwhelming evidence.”
AI Techno-Authoritarianism Identity Crisis
While sovereign AI is also considered a way of reinforcing national identity and security, it raises significant ethical issues. With nearly 47 million users on DeepSeek and 35 million on Grok, the influence of LLMs on public sentiment cannot be disputed.
Bill Gates recently warned that AI digital infrastructure could further polarize politics. In an interview with Handelsblatt Disrupt, Gates emphasized the dangers of employing AI to “exclude opposing views” and urged a transparent, open regulation process.
However, economic AI regulation itself risks becoming a tool of control. In authoritarian states like China, LLMs and facial recognition are being used to extend censorship and monitor dissent. This approach will extend censorship… further strengthening the authoritarian grip.
A study by the University of Copenhagen also found that some LLMs echo American cultural biases, which could “reaffirm cultural hegemony” rather than promote AI localization and global understanding.
As AI state suppression tools become embedded in education, governance, and public life, experts warn of a “programmable illusion” being shaped by unseen algorithms. Without international cooperation and shared ethical standards, the digital world risks becoming fragmented—where truth is redefined by code.
Inside Telecom provides you with an extensive list of content covering all aspects of the tech industry. Keep an eye on our Intelligent Tech sections to stay informed and up-to-date with our daily articles.