Unverified50% confidenceOpinionExact time
Off-the-shelf large language models (such as ChatGPT and Claude) are not safe for children, as their native content moderation mechanisms are designed for general adult scenarios.
1
Sources
50%
Confidence
Long-term
Relevance
9/7/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCross-lingual jailbreaking exploits the asymmetry between a model's concept space and its language space, as the learning of safety-refusal behavior is highly language-specific.67% similarUnverified确定性防护无法处理自然语言层面的隐晦风险，是其主要局限66% similarUnverified论文提出为防止大语言模型声称拥有意识而进行的安全微调会意外改变模型对'心智'的整体表征，扭曲其对人类信仰与价值观的建模64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/872888API
curl https://kongchang.com/api/v1/knowledge/claims/872888MCP
get_claim(id=872888)