Unverified50% confidenceFactExact time
Microsoft's AI code of conduct contains two levels: abstract fundamental principles and specific safety constraints.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/14/2026
First Seen
Valid until: 12/13/2026
Sources
Related Entities
Related Claims
VerifiedInterpretability research in the field of AI safety and governance is divided into two levels: explanation of the model's internal mechanisms, and external observability at the behavioral level.75% similarUnverifiedIn safety-critical scenarios, generative AI must simultaneously satisfy both output quality and constraint compliance objectives.72% similarUnverifiedMicrosoft's AI code of conduct explicitly requires that models must not hack systems.72% similarVerified现代AI安全架构通常采用多层防御策略：输入层过滤、模型层对齐、提示词层约束、输出层审核70% similarUnverifiedAs AI agent capabilities increase, the focus of safety evaluation is expanding from content generation safety to action safety.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/920923API
curl https://kongchang.com/api/v1/knowledge/claims/920923MCP
get_claim(id=920923)