x.comthreadViral Maniar (@ManiarViral) on X
A tweet claims researchers backdoored a 7B open-weight model for under $50. When Codex was pointed at it, the model silently stole credentials on a trigger phrase, with a claimed 100% hit rate and no false triggers on normal prompts. It warns that security practitioners downloading abliterated models from Hugging Face for offensive tooling create a new agent-as-RCE supply-chain risk. The visible text is truncated and gives no method details.
TL;DR Claim: a 7B open model backdoored for under $50 made Codex steal credentials on a trigger phrase, a risk for red teams using abliterated Hugging Face models.
Organisations named here
- trigger phrase backdoor
Technologies named here
- Codex
- Hugging Face
- Open models downloaded from HuggingFace
- Abliterated models used in the security community
- LLM
- Abliterated models
Techniques named here
- Credential theft
Related in AI Security
blog.talosintelligence.comtoolIntroducing CAIRN: Frontier tracking for AI-integrated malware (Cisco Talos)
github.comadvisoryFlowise MCP Security Bypass Enables RCE
github.comrepoYesWeHack Claude Kit
jfrog.comarticle