Unverified50% confidenceOpinionExact time
Photorealistic-style models come with ethical risks such as deepfakes.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
Unverified提示注入攻击可通过角色扮演（如声称自己是小说编剧）诱导大模型生成违规内容62% similarUnverified媒介学者约书亚·格林伯格将Bourdain的风格归纳为脆弱性叙事：叙述者主动暴露自身局限反而建立更高可信度59% similarUnverified凯文·米特尼克在其著作《欺骗的艺术》中系统记录了社会工程学的攻击框架58% similarUnverifiedAI-generated books on topics like health, investing, or parenting that contain misinformation can cause real-world harm to readers56% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/913851API
curl https://kongchang.com/api/v1/knowledge/claims/913851MCP
get_claim(id=913851)