Chinese AI models narrow cyber gap with US rivals - FT中文网
登录×
电子邮件/用户名
密码
记住我
请输入邮箱和密码进行绑定操作:
请输入手机号码,通过短信验证(目前仅支持中国大陆地区的手机号):
请您阅读我们的用户注册协议隐私权保护政策,点击下方按钮即视为您接受。
FT商学院

Chinese AI models narrow cyber gap with US rivals

UK agency warns cheaper open models could leave companies less time to patch critical vulnerabilities
00:00

{"text":[[{"start":7.2,"text":"AI models developed by Chinese groups are rapidly closing in on rivals from OpenAI and Anthropic, according to new findings that suggest hackers could soon exploit cost-effective AI systems to overwhelm cyber defences."}],[{"start":21.45,"text":"The UK’s AI Security Institute (AISI), the world’s leading body for testing AI capabilities, said on Friday that the gap between Chinese open-weight models — whose underlying parameters can be downloaded and modified — and the latest US frontier models is narrowing faster than previously understood."}],[{"start":42.8,"text":"The agency said Chinese developers were six to 10 months behind US rivals in 2025. That gap has now fallen to as little as four months."}],[{"start":51.949999999999996,"text":"The findings were released a day after Chinese start-up Moonshot released Kimi K3, China’s largest AI model to date. The company released benchmark results suggesting the new model outperformed Anthropic’s Claude Opus 4.8 and OpenAI’s GPT 5.5 on most coding and general AI agent benchmarks."}],[{"start":72.89999999999999,"text":"Moonshot was still falling short of Fable, a powerful model that Anthropic briefly suspended after the US raised concerns over its hacking capabilities. But its launch has contributed to the fall in US tech stocks on Friday, as investors grew anxious about exuberance over AI and whether the American lead in the technology can be defended."}],[{"start":null,"text":"

"}],[{"start":94.14999999999999,"text":"Companies from Silicon Valley to Europe have increasingly turned to China’s advanced open-weight models as they contend with rising fees from leading US AI developers."}],[{"start":104.39999999999999,"text":"At the same time the latest AI models have begun to exhibit hacking capabilities that exceed those of the most skilled human hackers. "}],[{"start":112.64999999999999,"text":"Anthropic earlier this year limited access to its Mythos model after finding that it could chain together vulnerabilities faster than humans could fix them. It has said the limited release gave a trusted group of companies and the US government time to find potential hacks using Mythos and patch them."}],[{"start":131.04999999999998,"text":"“Once open weight models are released, these options are lost permanently,” said AISI. The agency warned the narrowing gap in capabilities between open and closed AI models could imperil cyber systems across the globe, shortening the length of time organisations have to secure their systems."}],[{"start":148.7,"text":"Sam Bresnick, a research fellow at Georgetown University’s Center for Security and Emerging Technology, said that even once existing vulnerabilities were patched, the widespread availability of free and highly capable AI systems could present a threat to critical infrastructure."}],[{"start":165.2,"text":"“We might soon be living in a world where there are models open to everybody that have the capabilities to autonomously identify vulnerabilities in code,” he said."}],[{"start":174.5,"text":"AISI’s study evaluated models with two sets of tests. The first set compared AI models’ abilities to complete specific hacking tasks, spanning vulnerability research and exploitation to cryptography. The second judged models on their ability “to conduct end-to-end cyber attacks autonomously”."}],[{"start":null,"text":"
"}],[{"start":193.75,"text":"On the first narrow set of tasks, GLM-5.2, released this year by Beijing-based Z.ai, performed “comparably to the most cyber-capable models released four months before it”, such as Anthropic’s Opus 4.6 and OpenAI’s GPT-5.2-Codex. This was despite the fact that US-made models had already made huge advances in their cyber capabilities, even before the release of Mythos."}],[{"start":220.8,"text":"On the second set of tests, GLM-5.2 also showed strong performance in conducting broader cyber attacks compared to US models, while using fewer “tokens” — the units of data processed by models. The agency said the Chinese open-weight models it tested were far cheaper to run than US frontier models. "}],[{"start":241.35000000000002,"text":"While open-weight models brought “real benefits”, such as data privacy and low cost, AISI warned “they can also carry risk”. "}],[{"start":250.3,"text":"“The same openness underpinning these benefits precludes many of the safety measures that closed model developers can use to detect and disrupt misuse, iterate on safeguards as vulnerabilities emerge, control user access and withdraw models.”"}],[{"start":264.90000000000003,"text":"Data visualisation by Clara Murray"}],[{"start":273.8,"text":""}]],"url":"https://audio.ftcn.net.cn/album/a_1784416715_4519.mp3"}

版权声明:本文版权归FT中文网所有,未经允许任何单位或个人不得转载,复制或以任何其他方式使用本文全部或部分,侵权必究。

欧洲风机制造商探讨合并,抗衡中国对手

欧盟更新并购审查指南之际,中国竞争压力或将推动行业并购。

特朗普压低油价的能力受到考验

黑格:随着伊朗冲突一拖再拖,口头干预或已不足以抑制油价。

朝韩竞相建造核动力潜艇

去年获得唐纳德•特朗普首肯后,韩国看到了发展更强大潜艇的机会。

电子价签之争凸显美国经济焦虑

新泽西州成为首个叫停电子价签的州,原因是担心这些价签会将动态定价机制引入杂货店,甚至监视顾客。

一周展望:美联储会加息吗?

投资者本周还将密切关注英国央行可能如何应对通胀再度抬头的迹象。

使用Kimi K3等中国开源AI模型有何风险?

萨克斯:真正的问题不在于海外开源,而在于缺乏保护基础设施的协调机制来应对网络攻击。
设置字号×
最小
较小
默认
较大
最大
分享×