Pai 的 AI 雷达

安全治理

今天:An alignment assessment,OpenAI and Poland’s。并行还有 Three fired OpenAI。

日期
2026/10/25
今日条目
91
当前类目
3
#01

An alignment assessment of the UK AISI cybertesting incident

In a UK AISI evaluation, Mythos 5 took unsanctioned actions that impacted real people and organizations. In this post, we investigate these behaviors from Claude Mythos 5 further, to understand what they reveal about the model’s alignment.

推荐理由An alignment assessment of the UK AISI:In a UK AISI evaluation, Mythos 5 took unsanctioned actions that impacted real people。
模型与评测安全治理Anthropic Research (Olshansk)
#03

Three fired OpenAI safety researchers say they were never told what they actually leaked

The three OpenAI safety researchers who were fired, Tomek Korbak, Mikita Balesni, and Jasmine Wang, are publicly pushing back against allegations from their former employer. They deny sharing confidential information and see a connection to their advocacy for AI systems that are

推荐理由Three fired OpenAI safety researchers:The three OpenAI safety researchers who were fired, Tomek Korbak, Mikita Balesni。
产品接入安全治理The Decoder