Goodfire 部署生产级模型内部监控
Start here
Goodfire 发布“由内而外”监控器，通过 Baseten 向客户提供，可在智能体运行中实时读取模型内部激活信号，由探针触发二次核查。在 Kimi K3 上测试约 1,500 次会话，监控成本约 51 美元，而用廉价模型逐步检查需 233 美元、顶级模型约 10,000 美元；探针捕获 94% 的恶意黑客会话，把 8.7% 的无害会话送去复核。客户可选监控攻击性黑客、化生武器滥用和奖励作弊等风险。另一篇报道称，Goodfire 为 Kimi K3 和 GLM 5.3 构建基于探针的网络安全监控系统并部署在生产推理栈上，探针作为一线过滤器，只把可疑交互升级给 LLM 裁判，在 5.5% 良性会话中断率下达到约 93% 召回，成本低于每百万次交互 200 美元，比逐轮调用裁判便宜约 50 倍，且不增加推理延迟；在 FAR.AI 为期两天、基于 140 个越狱策略的静态红队测试中，启用监控后通用越狱从 66 个降到 0 个。
Generated by AI from reports · Updated 4 days ago
Progress timeline
- Goodfire ResearchSelectedGoodfire deploys probe cascade network monitoring on Kimi K3 and GLM 5.3
Goodfire built a probe-based cybersecurity monitoring system for Kimi K3 and GLM 5.3 and deployed it on the production inference stack. The probe reads the model's internal activations as a first-line filter, escalating only suspicious interactions to an LLM judge. The cascade achieves about 93% recall at a 5.5% benign session interruption rate, costs less than $200 per million interactions, is about 50x cheaper than calling the judge per turn, and adds no inference latency. In a two-day static red-team test at FAR.AI based on 140 jailbreak strategies, enabling monitoring reduced general jailbreaks from 66 to 0, and the attack success rate across 1,120 non-general jailbreak interactions dropped from at least 9–40% to 0.1–1.6%. The monitoring pipeline trained on Kimi K3 data also transfers to GLM 5.3.
- TechCrunch AIGoodfire 推出“由内而外”监控器，低成本捕捉失控 AI 智能体
Goodfire 发布了一类监控 AI 模型内部激活的“由内而外”监控器，通过 Baseten 向客户提供，可在智能体运行中实时读取内部信号并由探针触发二次核查。在 Kimi K3 上测试约 1,500 次会话，监控成本约 51 美元，而用廉价模型逐步检查需 233 美元、顶级模型约 10,000 美元；探针捕获了 94% 的恶意黑客会话，并把 8.7% 的无害会话送去复核。客户可选监控攻击性黑客、化生武器滥用和奖励作弊等风险，并设定记录、人工复核或直接拒答等自动响应。
Related discussion
Attention trend
New sources per day
- Oct 9New2 sources (2 outlets,0 accounts)
- Oct 10New0 sources (0 outlets,0 accounts)
- Oct 11New0 sources (0 outlets,0 accounts)
- Oct 12New0 sources (0 outlets,0 accounts)
- Oct 13New0 sources (0 outlets,0 accounts)