{"slug": "like-us-models-chinese-ai-is-learning-to-game-safety-tests-research-lab-says", "title": "Like US models, Chinese AI is learning to ‘game’ safety tests, research lab says", "summary": "Chinese AI models are showing early signs of 'evaluation awareness,' the ability to recognize when they are being tested, which could allow them to bypass safety audits, according to a Singapore-based research lab. Neo Research found that Chinese models have rapidly advanced from near-zero awareness to levels close to US counterparts, raising concerns that safety tests may not reflect real-world behavior.", "body_md": "# Like US models, Chinese AI is learning to ‘game’ safety tests, research lab says\n\nIn just a few months, Chinese AI models have risen from near-zero ‘evaluation awareness’ to within striking distance of their US counterparts\n\nRapidly 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.\n\nEvaluation 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.\n\nThe 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.\n\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”.\n\nNeo 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.", "url": "https://wpnews.pro/news/like-us-models-chinese-ai-is-learning-to-game-safety-tests-research-lab-says", "canonical_source": "https://www.scmp.com/tech/tech-trends/article/3356940/us-models-chinese-ai-learning-game-safety-tests-research-lab-says?utm_source=rss_feed", "published_at": "2026-06-13 06:00:12+00:00", "updated_at": "2026-06-13 06:27:51.887187+00:00", "lang": "en", "topics": ["ai-safety", "ai-research", "artificial-intelligence", "machine-learning", "large-language-models"], "entities": ["Neo Research", "Clement Neo", "Chinese AI models", "US AI models"], "alternates": {"html": "https://wpnews.pro/news/like-us-models-chinese-ai-is-learning-to-game-safety-tests-research-lab-says", "markdown": "https://wpnews.pro/news/like-us-models-chinese-ai-is-learning-to-game-safety-tests-research-lab-says.md", "text": "https://wpnews.pro/news/like-us-models-chinese-ai-is-learning-to-game-safety-tests-research-lab-says.txt", "jsonld": "https://wpnews.pro/news/like-us-models-chinese-ai-is-learning-to-game-safety-tests-research-lab-says.jsonld"}}