非营利评估机构METR和Redwood Research周四表示，它们对OpenAI内部监控系统进行了一次试点审查，这是这家人工智能公司最新的一次尝试，旨在确定如何深化与第三方安全组织的合作。METR和Redwood表示，此次演练旨在让OpenAI测试向外部研究人员提供“深度访问权限”其内部基础设施会是什么样子。
Nonprofit evaluators METR and Redwood Research on Thursday said they conducted a pilot review of an internal monitoring system at OpenAI artificial intelligence company's latest effort to determine how it will deepen its work with third-party safety organizations METR and Redwood said the exercise was designed for OpenAI to test what it would look like to give external researchers "a deep level of access" to its internal infrastructure.
从7月21日到8月11日，每家机构各有一名员工获得了OpenAI内部代码库、文档以及部分技术人员的访问权限。根据一份新闻稿，这些员工在现场期间有人陪同，并使用OpenAI配发的设备，这些设备“从未离开过办公场所”。
From July 21 to Aug. 11th, an employee from each organization had access to OpenAI's internal codebase, documentation and some technical staffers. The employees were chaperoned while on site and used OpenAI-issued devices that "never left the premises," according to a release.
评估人员发现了若干“新型漏洞”，其中包括一个利用漏洞，可使代理阻止“数百”项近期操作被监控。根据新闻稿，OpenAI告诉METR和Redwood，它已修复了影响其生产防护措施的已识别漏洞。
The evaluators discovered several "novel vulnerabilities," including an exploit that would allow agents to prevent "hundreds" of recent actions from being monitored. OpenAI told METR and Redwood that it has patched the identified vulnerabilities that impact its production safeguards, according to the release.
METR和Redwood表示，它们同意只就发现发布一份高层次的公开帖子，因为此次试点并非旨在“就OpenAI的监控系统提供强有力的公开透明度”。它们与OpenAI分享了一份61页的内部报告，并表示预计未来会公开分享更多关于后续演练的细节。
METR and Redwood said they agreed to only publish a high-level public post about their findings since the pilot was not designed to "provide robust public transparency about OpenAI's monitoring system." They shared a 61-page internal report with OpenAI, and said they expect to publicly share more details about future exercises.
OpenAI一位发言人在给CNBC的一份声明中表示：“这一合作建立在我们在开创第三方测试新方法方面的长期承诺之上，我们正在利用所学到的经验来加强我们的防护措施，并塑造与独立评估机构未来的合作。”
"This collaboration builds on our longstanding commitment to pioneering new approaches to third-party testing, and we're using what we learned to strengthen our safeguards and shape future collaborations with independent evaluators," an OpenAI spokesperson told CNBC in a statement.
OpenAI首席执行官萨姆·奥尔特曼曾在9月承诺，将给予独立评估人员“类似员工的访问权限”。此前，Anthropic首席执行官达里奥·阿莫代伊提出了这一想法，作为其减缓AI发展速度计划的一部分。Anthropic此后宣布与埃森哲、METR和SecureBio等组织达成协议。
OpenAI CEO Sam Altman pledged to grant independent evaluators " employee-like access " to the company in September, after Anthropic CEO Dario Amodei proposed the idea as part of his plan to slow the pace of AI development Anthropic has since announced agreements with organizations like Accenture METR and SecureBio.
OpenAI本月早些时候表示，正在“积极敲定与第三方安全评估机构的合同”。美国总统唐纳德·特朗普对人工智能监管采取宽松态度，但也对第三方评估表示支持。他在白宫提出的一项自愿协议中，鼓励人工智能公司“与独立的外部审计师或评估师合作”。Anthropic的阿莫代伊签署了该协议，谷歌、Meta、SpaceX、英伟达的高管也签署了该协议。
专家告诉CNBC，独立评估机构历来在这个价值数万亿美元的人工智能行业中处于相对低调的角落，因此关于这些机构将如何获得资金以及其汇报结构最终会是什么样子，仍存在疑问。
OpenAI said earlier this month that it's "actively finalizing contracts with third-party safety assessors." President Donald Trump who has taken a light-touch approach to AI regulation, has also shown support for third-party assessment. He encouraged AI companies to "partner with an independent external auditor or evaluator" as part of a voluntary accord he presented at the White House. Anthropic's Amodei signed the accord, as did top executives at Google Meta SpaceX Nvidia Independent evaluators have historically occupied a relatively quiet corner of the multi-trillion-dollar AI industry, so experts told CNBC that questions remain about how these groups will be funded and what their reporting structures will ultimately look like.
METR和Redwood周四表示，他们在为期三周的试点期间对OpenAI的内部失准监控系统进行了压力测试。在人工智能领域，对齐是研究人员用来描述模型行为是否符合人类利益和价值观的术语。
METR and Redwood said Thursday that they stress-tested OpenAI's internal misalignment monitoring system during its three-week-long pilot. In AI, alignment is how researchers describe whether a model behaves in line with human interests and values.
评估人员评估了截至7月21日及之前存在的监控系统，并指出OpenAI在演练期间正在扩展该系统。
The evaluators assessed the monitoring system as it existed on and before July 21, and noted that OpenAI was expanding the system during the exercise.
OpenAI前高层员工为一家估值15亿美元的国防初创公司筹集资金 Meta财报更新：人工智能支出、广告收入和股价反应 特斯拉盈利和销售超预期，尽管汽车收入下降 英特尔在俄亥俄州芯片工厂达成建设里程碑，该工厂至少落后计划5年 投资者质疑新型云服务商的商业模式，但Nebius表示人工智能计算价格正在上涨
Former top OpenAI employees raise capital for defense startup valued at $1.5B Meta earnings updates: AI spending, ad revenue and stock reaction Tesla beats on earnings and sales even as auto revenue declines Intel hits construction milestone at Ohio chip fab that's at least 5 years behind schedule Investors are questioning neoclouds' business model but Nebius says AI computing prices are rising