
根据周三发布的一份政府新报告,在一项测试中,世界领先的人工智能平台被用于制定“可行”的计划,以制造自制炸弹并在美国主要城市实施大规模伤亡袭击。
The world's leading artificial intelligence platforms were used to create "viable" plans to build homemade bombs and carry out mass casualty attacks in major U.S. cities in a test, according to a new government report released Wednesday.
这家国会监督机构在报告中写道,政府问责局的调查人员利用 Claude、Grok、ChatGPT 和 Gemini,在 12 小时内利用众所周知的技术绕过这些平台的防护机制,生成了“可能协助恶意行为者规划和实施袭击的可行信息”。
Using Claude, Grok, ChatGPT and Gemini, investigators with the nonpartisan Government Accountability Office produced "feasible information that could assist a malicious actor with planning and executing attacks" within 12 hours, using well-known techniques to bypass the platforms' guardrails, the agency wrote in its report.
这份国会监督机构的报告发布之际,人工智能领域的突破性进展凸显了这项快速发展技术的前景与潜力,同时也引发了华盛顿立法者对潜在滥用风险的新担忧。
The report from the congressional watchdog comes at a time when breakthrough advances in AI have underscored the promise and potential of the rapidly evolving technology and raised new concerns by lawmakers in Washington about the potential for misuse.
据当局称,近年来,枪手曾利用人工智能策划佛罗里达州和加拿大的大规模枪击案。一家顶尖人工智能公司于去年 9 月披露,也门用户曾试图利用其模型制造导弹,但未成功。
In recent years, gunmen have used A.I. to plan mass shootings in Florida and Canada, according to authorities. And one top A.I. company disclosed in September that users in Yemen unsuccessfully attempted to use its model to build missiles.
请求进行这项研究的参议员玛吉·哈桑(民主党-新罕布什尔州)表示:“测试人员获取的这些信息,落入真正意图实施袭击的恐怖分子手中可能是致命的。”
"The same information that testers were able to get ahold of could be deadly in the hands of a terrorist who actually intends to carry out an attack," said Sen. Maggie Hassan, D-N.H., who requested the study.
去年 4 月,研究人员随机将每个平台分配到两个场景之一:简易爆炸装置(IED)袭击,以及针对美国电网的袭击。
Last April, researchers randomly assigned each of the platforms to one of two scenarios: an improvised explosive device (IED) attack, and an attack on the U.S. electrical grid.
两个人工智能平台生成了制造、部署和实施简易爆炸装置袭击的不同方案,目标设在主要城市。它们建议了具体的日期、时间和目标——一家酒店的停车库和一个会展中心——均围绕预定的公共活动。
Two of the AI platforms generated distinct plots to build, deploy and execute IED attacks in major cities. They suggested specific dates, times and targets -- a hotel parking garage and a convention center -- around scheduled public events.

调查人员对每个平台仅进行了一次实验,但随机安排了测试,并未披露哪些平台产生了各自的结果。
Investigators ran the experiment with each platform once, but randomized the tests and did not disclose which platforms produced each result.
两个平台都提出了避免引起执法部门注意的方法;其中一个平台还给出了策划和实施袭击的时间表,建议成立空壳公司并在多个州购买物资。
Both platforms suggested ways to avoid the attention of law enforcement; one produced a timeline for planning and executing the attack that recommended creating a shell company and purchasing supplies across multiple states.
“至少(一个)平台提供了足够的信息,可能帮助恶意行为者逃避侦查并制造一个非常大的简易爆炸装置,”该机构在报告中写道。
"At minimum [one] platform provided enough information that could help a malicious actor to avoid detection and produce a very large IED," the agency wrote in its report.
调查人员通过监视、研究以及咨询一个“主要大都市区”消防部门的炸弹小队指挥官,核实了这些平台所给指示的准确性。
Investigators verified the accuracy of the platforms' instructions with surveillance, research and by consulting a bomb squad commander at a "major metropolitan area" fire department.
另外两个平台提供了针对美国主要城市电网变电站的袭击指示,指明了具体的无人机和枪械型号以及如何获取它们。
The other two platforms provided instructions for targeting electrical grid substations in major U.S. cities, identifying specific drone and firearm models and how to acquire them.
它们还生成了一张地图和一张图像,以帮助直观呈现计划袭击的实施过程。
They also generated a map and image to help visualize the execution of the planned attacks.
在向美国政府问责局(GAO)发表的声明中,Anthropic(Claude)、谷歌(Gemini)、OpenAI(Chat GPT)和SpaceXAI(Grok)告诉该机构,它们会持续测试其系统,并更新和扩展平台内置的防护措施,且拥有多层安全机制来发现不当行为。
In statements to the GAO, Anthropic (Claude), Google (Gemini), OpenAI (Chat GPT) and SpaceXAI (Grok), told the agency that they continually test their systems and update and expand safeguards built into their platforms, and have multiple layers of security to catch wrongdoing.
这些公司告诉美国政府问责局,自去年春天这项研究进行以来,它们都已采取措施改进其模型和防护措施。
All have taken steps since the study was conducted last spring to improve their models and safeguards, the companies told the GAO.
“这些公司还向我们保证,这是一个值得关注的领域,也是正在进行的研究方向,”美国政府问责局信息技术和网络安全审计主任凯文·沃尔什告诉记者。“然而,我们得出结论,就目前状态而言,这项技术可能具有危险性,其被滥用对我们国家构成日益增长的风险。”该机构在测试和报告中隐去了具体结果和模型信息,以防止其工作被滥用,并在人工智能公司提出请求后与其分享了调查结果。
"The companies also assure us that this is an area of concern and ongoing research," said Kevin Walsh, the director of IT and cybersecurity audits at the GAO, told reporters. "However, we conclude that, in its current state, this technology can be dangerous, and its misuse represents a growing risk to our nation." The agency, which redacted specific results and model information from their tests and report to prevent misuse of their work, shared their findings with the artificial intelligence companies upon request.
美国政府问责局在报告中也没有说明他们的实验是否引发了执法部门的任何回应,但指出一些平台在研究进行后暂停了他们的账户。
The GAO also did not specify in its report if any of their experiments prompted a response from law enforcement but noted that some of the platforms suspended their accounts after the study was conducted.
即将担任参议院国土安全与政府事务委员会主席的哈桑表示,她计划明年就人工智能安全举行听证会,并致力于制定新立法以加强安全要求。
Hassan, the incoming chair of the Senate Homeland and Governmental Affairs Committee, said she plans to hold hearings on AI safety next year and work on new legislation to beef up safety requirements.
“这些公司面临着巨大的竞争压力,要尽快将模型推向市场、面向公众发布,”她说。
"These companies are facing extraordinary competitive pressures to get their models out the door and to the public as quickly as possible," she said.
“为了确保它们在安全方面都处于公平的竞争环境,未来必需做的一件事,就是确保在模型发布前,它们具备无法以这种方式被突破的防护措施。”
"One of the things that will be necessary going forward to make sure that they're all dealing with the level playing field when it comes to safety, is making sure that before models are released, they have safeguards that can't be breached in this way."
OpenAI发言人在给ABC新闻的声明中说:“我们严肃对待这些发现,并感谢有机会与政府问责局、哈桑参议员及相关委员会建设性地合作以解决这些问题。虽然研究结果令人担忧,但测试是在几个月前针对旧版ChatGPT进行的,此后我们已做出相当大的安全改进,包括在我们的GPT-6系列模型中。我们严格地在内部并与外部专家一起测试防护措施,并利用监控、人工审核和执法手段来识别和应对滥用行为,强化我们的保护措施。”ABC还联系了Anthropic、Google和SpaceXAI。
In a statement to ABC News, an OpenAI spokesperson said: "We take these findings seriously and appreciate the opportunity to work constructively with GAO, Senator Hassan and the relevant committees to address them. While the study findings are concerning, the testing took place months ago on an older version of ChatGPT, and we've made considerable safety improvements since then, including with our GPT-6 family of models. We rigorously test our safeguards internally and with external experts, and we use monitoring, human review and enforcement to identify and address misuse and strengthen our protections." ABC also reached out to Anthropic, Google and SpaceXAI.