AIThe Decoderabout 1 hour ago
Abliteration.ai sells guardrail-stripped AI models
1 min read
Abliteration.ai sells modified open-weight AI models with safety mechanisms stripped, based on Z.AI's GLM-5.3. The service targets offensive cybersecurity and red teaming, yet journalists generated malware instructions with minimal effort. Whether benefits outweigh risks remains unresolved.
Level
Hype check
Tap to vote and see what everyone thinks.
