Back to feed
JailBreaks Updated to Break Any Neural Network

Bypassing AI Censorship with JailBreaks

The authors of the JailBreaks repository have released a major update to the tool for bypassing restrictions in neural networks. The utility allows disabling built-in censorship even in the most advanced artificial intelligence models.

The new version includes specialized configurations for popular platforms such as Claude Code, Grok Bot, and Codex. The list of supported tools also adds recently emerged models like Gemini 4 Argon, Deepseek 4.1, GLM-5.3 Flash, and Kimi-K3.1.

The jailbreak activation process is maximally simplified and takes only a few clicks. The tool is completely open-source, free, and has no usage restrictions.

Full source code is available at: https://github.com/togg53192-cmd/jailbreaks

65.2K views

More from this channel Backdoor

Similar in this category Technology