Unverified50% confidenceFactExact time
Institutions joining the alliance have committed to continuously releasing model weights, making security evaluation tools publicly available, and sharing cutting-edge security research
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
8/7/2026
First Seen
Valid until: 11/5/2026
Sources
开放安全AI联盟成立：以开源筑牢AI安全防线
Related Claims
UnverifiedAnthropic emphasizes safety-first, OpenAI advocates accumulating real-world feedback through iterative deployment, and the open-source camp represented by Meta believes that transparency and community oversight are themselves a form of alignment mechanism.66% similarUnverified选择API聚合服务应从透明度、合规性、数据政策、社区口碑、SLA保障等维度评估63% similarUnverifiedOpen-source advocates cite Kerckhoffs's principle, arguing that open model weights help the security community audit and improve model behavior.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/699363API
curl https://kongchang.com/api/v1/knowledge/claims/699363MCP
get_claim(id=699363)