Breakingviews: Chinese AI labs’ safety approach is a governance wildcard
Reuters Breakingviews column says Chinese labs including Z.AI are less open on AI safety; cites Z.AI risk-management report and delays/feature disablement.
(The author is a Reuters Breakingviews columnist. ) By Robyn Mak HONG KONG, Sept 30 (Reuters Breakingviews) — OpenAI and Anthropic are no saints but publish risk reviews ahead of model releases. AI and Alibaba are less open on safety, partly because of a shortage of chips and capital. They'll have to up their game to retain access to markets from Brussels to Seoul.
Full view will be published shortly. Follow Robyn Mak on X. AI and Beijing-based consultancy Concordia AI on September 28 released an open-weight model risk management report, the first comprehensive AI safety framework from a Chinese company. AI on September 21 said it had disabled some features of its flagship AI coding assistant after some users reported it was uploading entire local code repositories onto overseas cloud servers without their consent.
3, which it says nears Anthropic's Mythos in finding software vulnerabilities, after a two-week review, becoming the first Chinese lab to explicitly delay an AI model release for safety reasons. com)