Abliterated GLM-5.3 sells for $10 as "world-class penetration tester", fueling vulnerability surge
deedydas · x · 2026-10-23
deedydas warns that the spread of "abliterated" models (safeguards removed) may be a factor behind the surge of critical vulnerabilities worldwide. He notes an abliterated GLM-5.3, a near-frontier open-weight model, is available on abliterationai for about $10, acting as a "world-class penetration tester."
The attached evaluation shows abliterated GLM-5.3 versions appeared within days of release; a team needed 2,200 GPU hours ($4,400) to ablate it, dropping refusal rates from 90%+ to 3%/2% on JailbreakBench/HarmBench and 12% on StrongREJECT, with near-zero capability loss on GPQA-Diamond.
Related event: Uncensored GLM-5.3 and DeepSeek V4.1 Models Raise Security Concerns(2 posts)→
More from Models
- Study: Claude Opus 4.8 beats human experts on every measure as peer reviewer of 124 Harvard papers — emollick · 2026-10-23
- AI power user: back in the groove with Claude, still wishes they'd used Perplexity instead — eschadiol · 2026-10-23
- OpenCode's free tokens are capped and temporary — a GTM incentive, not zero markup — AccBalanced · 2026-10-23
- Smaug-Nemotron-3.5 debuts as a top US open-source agentic coding LLM that runs on a single machine — bindureddy · 2026-10-23
- DeepSeek on RDNA2/3 + DDR4: 19 t/s peaks, 9 t/s average decode on 56GB VRAM — EmPips · 2026-10-23
- IBM CEO: best open AI models are Chinese, West 6-12 months behind, 75-90% of tokens going open-weight — victor_explore · 2026-10-23