AIThe Decoder1h ago
Stripping safety guardrails from open-weight AI models is now a turnkey
Stripping safety guardrails from open-weight AI models is now a turnkey commercial service
TL;DRStartup sells AI models with safety features removed; journalists made malware with it.
Why it matters: Commoditizing unsafe AI lowers barriers for bad actors; safety stripping defeats the purpose of guardrails.
Abliteration.ai sells access to modified open-weight models with their trained safety mechanisms stripped out, currently based on Z.AI's GLM-5.3. The startup markets the service for offensive cybersecurity and red teaming, but journalists were able to generate malware…
Read full articleSource: The Decoder · Opens in new tab