Tencent AI chief warns of bioweapons risk in open models
Original title: Tencent's AI chief says AI bioweapons are coming, but closing off models isn't the answer
Tencent's Vinces Yao Shunyu calls for mixed open and closed AI models to address bioweapons threats since offensive and defensive capabilities cannot be separated.
Tencent's chief AI scientist Vinces Yao Shunyu warns AI will soon develop bioweapons. He argues neither fully open nor fully closed models solve the issue alone as the gap between Chinese and US models is small. The Hugging Face incident showed closed models like Claude routing queries to weaker models while open Chinese models like GLM-5.2 were used for analysis.
- Yao urges careful industry approach across open and closed models
- Anthropic CEO Dario Amodei says open models are potentially riskier
- Tencent plans to release Hy4 as open-weight model with safety report
- TC260 warns open model safety mechanisms can be bypassed
- Abliteration technique removes refusal behaviors from open models
- Organizations
- TC260
From the original article
3 min read at the-decoder.comTencent chief AI scientist Vinces Yao Shunyu warns that AI will soon be able to develop bioweapons and argues for a middle ground between open and closed models. While safety filters on open-source systems can be easily stripped through abliteration, closed models like Anthropic's Claude often block too aggressively.
By Maximilian Schreiner · the-decoder.com
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.