WIRED profiles the 'abliterators' stripping safety guardrails from open-weight AI models

willknight · x · 2026-11-11

A WIRED feature by Will Knight explores the "abliteration" subculture: people who strip refusal guardrails from open-weight models. Key points:

• Devon Thomas, a former Apple AI engineer, founded Abliteration AI to remove safeguards from top open-weight models and offer them to anyone.
• His motivation includes distrust of hosted models: data sharing concerns, models changing overnight, and the view that guardrails are censorship.
• The piece challenges the narrative that uncensored models mainly serve criminals—supporters cite uses like cyberdefense—while acknowledging risks like weapons guidance.
• A representative look at the ongoing clash between AI safety norms and the "free the models" movement.

Original post →

More from Models

• Spotify's Claude Code 90% token-saving trick put to a real test — aakashgupta · 2026-11-11
• Apple reportedly building Harbor app to turn Macs into private AI inference servers with Thunderbolt 5 clustering — BenBajarin · 2026-11-11
• Agent Arena: Gemini 4 Carbon Gets More Cost-Competitive as Tasks Get Harder — rseroter · 2026-11-11
• LiquidAI's 1.7B LFM3 hits SOTA on Multi-IF, beating 100x-larger rivals including ERNIE 5.0 — NielsRogge · 2026-11-11
• Fable 5.2 tops GDPx chip-design benchmark, but task authors see zero gains — ycombinator · 2026-11-11
• 1.3B open-source MiniCPM-OCR 1.0 tops Real5-OmniDocBench, beating GPT-5.2 — NielsRogge · 2026-11-11

Models channel →