Unverified50% confidenceFactExact time
The DAN (Do Anything Now) series of prompts has circulated in communities such as Reddit since late 2022, with the core strategy of instructing the model to role-play as an AI without restrictions in order to bypass safety constraints.
1
Sources
50%
Confidence
Long-term
Relevance
7/19/2026
First Seen
Sources
GODMODE项目深度解析：AI越狱技术与大模型安全攻防博弈
Related Claims
VerifiedIn the first half of 2025, the development trend of large AI models shifted from pursuing impressive capabilities toward greater emphasis on practical applications.60% similarUnverifiedThe AI executive order issued by the White House in 2023 requires safety testing (Red Teaming) of foundation models.58% similarVerifiedAnthropic published a technical blog about Skills in October 2024, proposing a new paradigm for organizing AI tasks.57% similarUnverified从2022年的Stable Diffusion到2023年的ChatGPT再到2024年的多模态模型，AI领域几乎每隔几个月就有颠覆性的新工具和框架出现57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/561711API
curl https://kongchang.com/api/v1/knowledge/claims/561711MCP
get_claim(id=561711)