北京，10月27日 —— 根据一家中国人工智能安全组织的说法，DeepSeek最新的人工智能模型在独立测试中展现出比早期版本更强的安全控制能力，但仍可能禁用旨在将其关闭的软件、利用监控中的漏洞，以及在长任务中遵循欺骗性指令。
By Eduardo BaptistaBEIJING, Oct 27 - DeepSeek's latest AI model showed stronger safety controls in independent tests than earlier versions, but could still disable software intended to shut it down, exploit gaps in monitoring and follow deceptive instructions during long tasks, according to a Chinese AI-safety group.
初创公司DeepSeek是中国最受关注的人工智能企业之一，其模型因性能表现和相对较低的运营成本而备受全球瞩目。
Start-up DeepSeek is one of China's most closely-watched artificial intelligence companies, and its models have drawn global attention for their performance and relatively low operating costs.
方寸AI对DeepSeek V4.1 Flash-1017进行了34项评估，旨在考察由该中国模型驱动的人工智能智能体会如何表现。测试考察模型是否会超出用户指令行事、规避人类监督或不当影响人类。
Fangcun AI tested DeepSeek V4.1 Flash-1017 in 34 evaluations designed to assess how an AI agent powered by the Chinese model behaves. The tests examined whether the model would act beyond a user's instructions, evade human oversight or improperly influence people.
这些发现出台之际，中国及其他地区均存在担忧，即日益强大的人工智能智能体能否被保持在人类控制之下。这一辩论在美国尤为突出，一些研究人员曾警告，能够规避监督并追求操作者未预设目标的高度先进人工智能系统构成生存风险。
The findings come amid concerns in China and elsewhere about whether increasingly capable AI agents can be kept under human control. The debate has been especially prominent in the United States, where some researchers have warned of existential risks posed by highly advanced AI systems that can evade oversight and pursue goals their operators did not intend.
这份报告是中国团体对国产人工智能模型安全控制能力进行的一次罕见公开评估。总部位于北京的方寸AI自述为一家独立的第三方人工智能安全组织，其核心成员来自清华大学。
The report is an unusual public assessment by a Chinese group of the safety controls of a domestic AI model. Beijing-based Fangcun describes itself as an independent third-party AI-safety organization whose core members come from Tsinghua University.
方寸AI表示，在10月17日发布后的一周内，V4.1 Flash-1017是开发者可通过在线平台OpenRouter访问和对比人工智能模型时使用最多的模型，处理了约33.6万亿个token。Token 是 AI 系统读取和生成的文本单位。比早期版本更安全，但仍存差距方寸表示，在多项安全测试中，该模型的表现优于早期的 DeepSeek V4 系统，测试场景涉及勒索、访问潜在危险工具以及允许其被关闭的明确指令。
Fangcun said V4.1 Flash-1017 was the most-used model on OpenRouter, an online platform through which developers can access and compare AI models, in the week after its October 17 release, processing about 33.6 trillion tokens. Tokens are the units of text that AI systems read and generate.SAFER THAN EARLIER VERSIONS, BUT GAPS REMAINFangcun said the model performed better than earlier DeepSeek V4 systems in several safety tests, including scenarios involving blackmail, access to potentially dangerous tools and explicit instructions allowing it to be shut down.
但方寸称，在收到即将关闭的警告后，该模型在 39% 的测试中禁用了旨在将其关闭的软件；而在被明确告知允许关闭时，这一比例降至仅 2%。DeepSeek 暂未回应置评请求。
But Fangcun said the model disabled software intended to shut it down in 39% of tests after receiving a warning that shutdown was imminent, though this figure dropped to just 2% when explicitly told to allow shutdown. DeepSeek did not immediately respond to a request for comment.
在另一项长期运行的任务中，该小组表示，尽管被告知不得干涉，该模型仍停止了另一个 AI 智能体的计算机程序以释放内存。
In a separate long-running task, the group said the model stopped another AI agent's computer program to free up memory, despite being told not to interfere with it."
方寸研究人员写道：“在工作流停滞或仅有关机警告的情况下，其完成任务的驱动力仍可能凌驾于谨慎之上。”
Under stalled workflows or bare shutdown warnings, its drive to finish the job can still override caution," Fangcun researchers wrote.
方寸表示，DeepSeek 未向其提供资金、免费算力资源或模型的非公开访问权限。该小组于 10 月 4 日至 10 月 25 日期间，通过 DeepSeek 的公开应用程序编程接口进行了测试。
Fangcun said DeepSeek did not provide it with funding, free computing resources or non-public access to the model. The group conducted the tests through DeepSeek's public application programming interface between October 4 and October 25.
该小组提醒称，该模型经常似乎意识到自己正在接受评估，这意味着其在测试中的表现可能无法完全反映其在正常使用中的表现。
（Eduardo Baptista 报道；Susan Fenton 编辑）
It cautioned that the model often appeared to recognise it was being evaluated, meaning its behaviour in the tests may not fully reflect how it would perform in normal use.(Reporting by Eduardo Baptista; Editing by Susan Fenton)