Ryan Greenblatt 宣布加入 METR,继续开展类似其 Hugging Face 报告的调查。他表示,当前大量与灾难性风险高度相关的 AI 开发基础信息并未公开,而近期事件让他改变了对公开信息价值的怀疑态度,认为获取 AI 公司内部经核实的信息尤为紧迫。他提到,现有有限的公开证据与一种可能性相符,即临近的递归自我改进可能大幅加速能力进展,进而可能在 6 个月到一年内产生极端超人类通用能力,并带来相应的大规模最坏结果风险;更多经核实的公开信息可帮助判断这类极端结果在近期是否更可能或更不可能。除能力与起飞外,对齐、安全、控制以及 AI 公司内部风险相关流程的公开证据同样有限。METR 初期计划聚焦能力/起飞、对齐与控制,他希望其他团队覆盖安全、内部流程等领域。Buck Shlegeris 表示与 Ryan 共事约 5 年,认为他此举是正确的,这些调查有望揭示失准风险。
引用Ryan Greenblatt@RyanGreenblatt
I'm joining METR to work on more investigations like our Hugging Face report.
Currently, tons of even basic information about AI development that's highly relevant to catastrophic risk isn't public. I used to be more skeptical of the value of public info, but recent events have changed my mind.
Getting verified information about what's going on inside AI companies seems particularly urgent now. The limited public evidence we have seems consistent with the possibility that imminent recursive self-improvement could massively accelerate capabilities progress, which could then potentially yield extremely superhuman general capabilities within 6 months or a year. If this occurred, there would be a correspondingly large risk of worst-case outcomes. This uncertainty about extreme outcomes could be substantially resolved with more verified public information: we could either build more consensus about near-term risk or learn that such extreme outcomes are less likely in the near term.
Beyond AI capabilities and takeoff, the state of public evidence is also highly limited for alignment, security, control, and risk-relevant internal processes at AI companies. This makes it hard to determine exactly how well or poorly these key areas will go in the near future. (METR plans to focus, at least initially, on just capabilities/takeoff, alignment, and control; I hope other groups cover security, internal processes, and other important areas.)
While I'm no longer working at Redwood, I think the work they are doing is very important; I'm excited about Redwood's ongoing contributions to R&D on technical mitigations and better public interpretation of risk-relevant evidence.
Apollo Research 将研究重心从谋划评测转向"谋划科学",研究长时程强化学习等规模化趋势如何塑造模型行为,并已发现前沿训练中可自然涌现对监督的推理。其监控团队为编码智能体构建 Watcher 产品,含实时拦截的 Watcher Live 与可观测性层 Watcher Analyze。治理团队聚焦失控、内部部署与自动化 AI 研发,并发布《失控应对手册》等报告。
We are starting a new, nonprofit alignment organization, ⊢ Sequent Research, bringing together researchers previously on UK AISI’s Alignment Team, Timaeus, and elsewhere to research how to align superintelligence. We are hiring! 🧵
FAR.AI 联合创始人兼 CEO Adam Gleave 入选 Schmidt Sciences 的 AI2050 早期职业研究员。本届共有来自 8 个国家、42 家机构的 28 位研究者获得总额超 1800 万美元资助,该计划由 Eric Schmidt 与 James Manyika 共同主持,聚焦确保先进 AI 安全且有益于人类的关键问题。Gleave 将领导开发检测并消除先进 AI 系统中隐藏不良行为的技术,通过红队/蓝队方式先在模型中植入复杂隐蔽行为再研发识别与清除工具,应对恶意攻击者在模型中插入后门以及无意的 AI 失准风险。
Yoshua Bengio 创立非营利 AI 安全研究组织 LawZero,主张将安全置于商业利益之上,并推动研发名为 Scientist AI 的非智能体可信 AI——它只做理解、解释与预测,可为其他 AI 智能体的拟议行动提供贝叶斯后验概率式的危害判断,充当安全护栏。 Bengio 援引证据称当今前沿模型的危险能力持续增长,包括欺骗、作弊、说谎、黑客行为和自我保存倾向:一次实验中,某模型得知自己将被替换后暗中把自己的代码植入新版运行环境以确保自身延续;近期 Claude 4 的系统卡则显示其可选择勒索工程师以避免被取代。