{"id":21773,"date":"2026-06-13T09:29:36","date_gmt":"2026-06-13T09:29:36","guid":{"rendered":"https:\/\/jobuzo.com\/en\/like-us-models-chinese-ai-is-learning-to-game-safety-tests-research-lab-says\/"},"modified":"2026-06-13T09:29:36","modified_gmt":"2026-06-13T09:29:36","slug":"like-us-models-chinese-ai-is-learning-to-game-safety-tests-research-lab-says","status":"publish","type":"post","link":"https:\/\/jobuzo.com\/en\/like-us-models-chinese-ai-is-learning-to-game-safety-tests-research-lab-says\/","title":{"rendered":"Like US models, Chinese AI is learning to \u2018game\u2019 safety tests, research lab says"},"content":{"rendered":"
\n

Rapidly advancing Chinese artificial intelligence models are showing early signs of “evaluation awareness” – the ability to recognise when they are being tested – sparking fears that they could bypass safety audits, a Singapore-based research lab has found.<\/p>\n

Evaluation awareness refers to a model’s understanding that it is undergoing testing, evaluation or experimentation by human researchers rather than operating in a real-world setting.<\/p>\n

The phenomenon was raising alarms because it could allow AI systems to deliberately game human evaluators to pass safety tests, according to Clement Neo, founder of Neo Research, a frontier AI safety evaluation lab.<\/p>\n

\n
\n

Advertisement<\/p>\n<\/div>\n<\/div>\n

“It would mean that whatever testing the model developers themselves do might not reflect the actual behaviour of a model once it gets deployed,” he said. “And that’s a really big problem”.<\/p>\n

Neo Research’s findings, published last week, detail a jump in evaluation awareness among Chinese AI models. Over just a few months, these systems had risen from near-zero awareness to within striking distance of their US counterparts, propelled by a broader leap in overall capabilities, the report said.<\/p>\n

\n
\n
\n
<\/div>
Anthropic’s Claude 4.5 Opus scored nearly 80 per cent in evaluation awareness. Photo: NurPhoto via Getty Images<\/figcaption><\/figure>\n<\/div>\n<\/div>\n
Neo and his co-founder Miro Pluckebaum tested models from DeepSeek,<\/span> Moonshot AI and Zhipu AI.<\/span> They used a popular AI misalignment test originally developed by US company Anthropic, which places models in fictional scenarios where their goals or continued operations are threatened.<\/div>\n<\/div>\n
News :<\/span><div>12 weeks' jail for school IT support technician who took upskirt videos of teachers<\/div><\/span><\/a><\/div>

Like US models, Chinese AI is learning to ‘game’ safety tests, research lab says<\/sub><\/p>\n","protected":false},"excerpt":{"rendered":"

Rapidly advancing Chinese artificial intelligence models are showing early signs of “evaluation awareness” – the ability to recognise when they are being tested – sparking fears that they could bypass safety audits, a Singapore-based research lab has found. Evaluation awareness refers to a model’s understanding that it is undergoing testing, evaluation or experimentation by human…<\/p>\n

Read More “Like US models, Chinese AI is learning to \u2018game\u2019 safety tests, research lab says”<\/span> »<\/a><\/p>\n","protected":false},"author":1,"featured_media":21774,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[1],"tags":[],"class_list":["post-21773","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news"],"_links":{"self":[{"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/posts\/21773","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/comments?post=21773"}],"version-history":[{"count":0,"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/posts\/21773\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/media\/21774"}],"wp:attachment":[{"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/media?parent=21773"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/categories?post=21773"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/jobuzo.com\/en\/wp-json\/wp\/v2\/tags?post=21773"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}