State propaganda in some Chinese AI models is only "a thin layer" that fine-tuning can strip away
The Decoder ·
Summary: Nous Research retrained Alibaba's Qwen3.8-27B to produce Arete-27B, largely stripping out its pro-Chinese bias. The propaganda rate dropped from 22 percent to less than one percent, while general capabilities remained intact in tests. The method relied on 53 million tokens of sup
Read the full story on The DecoderExplore more: Today in AI · Research papers · AI tools · Learn AI
Related stories
- Bioengineer.org · 7 Nov 2026Vignette Study Compares AI Models and Physicians in Tuberculosis Clinical Tasks
- BGR · 7 Nov 2026Who Owns DeepSeek, The Chinese AI Company?
- The Verge AI · 7 Nov 2026Why a new report on Anthropic automating AI development is alarming researchers
Anthropic is using AI to build more powerful AI. And it appears to be working remarkably well, with the company's models improving roughly twice as quickly as a year ago, with little indication of slowing down. That's according to a new report published Thursday by the AI safety
- Bioengineer.org · 7 Nov 2026Fine-Tuned Language Model Discovers Critical Assembly Rules for Commercial Aircraft Manufacturing
- UC Today · 7 Nov 2026Microsoft Ignite 2026: Will Copilot Prove Its Value?
- Seeking Alpha · 7 Nov 2026The humanoid robot race heats up after Agility Robotics begins to trade (AGLT:NASDAQ)
Who reported “State propaganda in some Chinese AI models is only "a thin layer" that…”?
The Decoder, on 7 Nov 2026. The full article is at the source link on this page.