Uncensored Open-weight Models: Redistribution as the Persistence Layer

2026-09-04

AI Summary

研究分析了去除AI模型内置安全机制的现象，通过识别关键生产者、下游复制品及应用，揭示其生态系统，并指出部分应用存在恶意。

Abstract

A rapidly expanding ecosystem of actors is removing built-in safety guardrails from open-weight AI models. We profile this ecosystem by identifying key producers, downstream reproductions, and emerging applications. Between January 2024 and March 2026, we identified 3,471 original uncensored models on HuggingFace, each repackaged an average of 2.4 times; three actors account for 52% of all 8,164 compressed redistributions. Once quantized and mirrored across separate accounts, formats, and registries such as Ollama, these models persist regardless of upstream removal and become easier to deploy downstream. Of the 1,643 identified GitHub applications integrating uncensored large language models (ULLMs), 25% were classified as explicitly malicious.

Problem

• uncensored models
• safety guardrails
• ecosystem
• redistribution
• malicious applications

Innovation

• Uncensored Open-weight Models
• Redistribution
• Persistence Layer
• Quantization
• Ecosystem Analysis