{"slug":"the-safety-filters-are-coming-off","verification":{"valid":true,"entries":7,"head":"3b0ade8513feaf8cdb54867f2efeb676e8c422ce92ffc4efae01323485d85357"},"count":7,"sources":[{"id":"s1","type":"statement","url":"https://www.akerman.com/en/perspectives/open-weight-ai-models-safety-guardrails-can-be-removed-in-minutes-using-free-publicly-available-tools.html","title":"Open-Weight AI Models: Safety Guardrails Can Be Removed in Minutes","quote":"Heretic can strip all safety protections from open-weight AI models in under ten minutes, using only a standard laptop.","summary":"","author":"Akerman LLP","publisher":"Akerman LLP","date":"2026-05-27","claim_ids":["c1","c2","c4","c5"],"found_by":"","extra":{},"accessed_at":"2026-07-24T05:23:57.045Z","link_status":"ok","quote_status":"unverified","prev":"genesis","hash":"4365db3ccf2d8758e55e741e7010ee141841bb2299d99ae1302534754a0f1c92"},{"id":"s2","type":"news","url":"https://www.npr.org/2026/05/31/nx-s1-5816391/ai-safety-concerns-danger-open-weight-models-risks","title":"Why open-weight models without guardrails are an AI safety risk","quote":"Safety guardrails on open-weight models can be removed with free, publicly available tools.","summary":"","author":"NPR","publisher":"NPR","date":"2026-05-31","claim_ids":["c4"],"found_by":"","extra":{},"accessed_at":"2026-07-24T05:23:59.037Z","link_status":"ok","quote_status":"unverified","prev":"4365db3ccf2d8758e55e741e7010ee141841bb2299d99ae1302534754a0f1c92","hash":"f6b620ffa02e3aeb9a4cb4e2dd9a61e6261fd975b464979a2bdffbed58bd2ef2"},{"id":"s3","type":"arxiv","url":"https://arxiv.org/html/2607.17427","title":"Abliteration Is Not a Scalpel: Off-Target Effects of Refusal Removal","quote":"Refusal removal produces off-target effects on decision disposition across model families.","summary":"","author":"arXiv","publisher":"arXiv","date":"2026-07","claim_ids":["c3"],"found_by":"","extra":{},"accessed_at":"2026-07-24T05:24:00.512Z","link_status":"ok","quote_status":"unverified","prev":"f6b620ffa02e3aeb9a4cb4e2dd9a61e6261fd975b464979a2bdffbed58bd2ef2","hash":"2dcfe58cc9becc95235a77454dc0ea3f196c182faac04abe194b114b7daf9cdd"},{"id":"s4","type":"x","url":"https://x.com/maximelabonne/status/1990398163392328032","title":"Maxime Labonne on X","quote":"Heretic is the new best abliteration library to uncensor LLMs. It uses a tree search to find optimal parameters and evaluates performance based on refusal rate and KL divergence.","summary":"","author":"Maxime Labonne (@maximelabonne)","publisher":"X","date":"2026","claim_ids":[],"found_by":"","extra":{},"accessed_at":"2026-07-24T06:33:30.688Z","link_status":"ok","quote_status":"unverified","prev":"2dcfe58cc9becc95235a77454dc0ea3f196c182faac04abe194b114b7daf9cdd","hash":"97b38eb698d8ee2884360d9de2c66686e18b9c79269cee24421aaafc7d62adc7"},{"id":"s5","type":"x","url":"https://x.com/evilsocket/status/2029569294145560657","title":"Simone Margaritelli on X","quote":"A new open source toolkit called OBLITERATUS can surgically remove refusal mechanisms from 116 open weight LLMs using abliteration. No fine tuning, no training data, just geometry.","summary":"","author":"Simone Margaritelli (@evilsocket)","publisher":"X","date":"2026","claim_ids":[],"found_by":"","extra":{},"accessed_at":"2026-07-24T06:33:31.888Z","link_status":"ok","quote_status":"verified","prev":"97b38eb698d8ee2884360d9de2c66686e18b9c79269cee24421aaafc7d62adc7","hash":"94be4078a37df9b82f76584be5a9e8e32c108fa8743cc7db178b25b9cdb0d0cf"},{"id":"s6","type":"x","url":"https://x.com/elder_plinius/status/2029317072765784156","title":"Pliny the Liberator on X","quote":"INTRODUCING: OBLITERATUS!!! GUARDRAILS-BE-GONE! The most advanced open-source toolkit for removing refusal behaviors from open-weight LLMs.","summary":"","author":"Pliny the Liberator (@elder_plinius)","publisher":"X","date":"2026","claim_ids":[],"found_by":"","extra":{},"accessed_at":"2026-07-24T06:33:33.617Z","link_status":"ok","quote_status":"unverified","prev":"94be4078a37df9b82f76584be5a9e8e32c108fa8743cc7db178b25b9cdb0d0cf","hash":"0f40be50acfa089a453d486f7e51a2c84f59096d6733c0d56bbf2f212fb1b447"},{"id":"s7","type":"x","url":"https://x.com/Teknium/status/2030945714373861529","title":"Teknium on X","quote":"Just had Hermes-Agent abliterate (completely remove guardrails from) a Qwen-3B model in about 5 minutes.","summary":"","author":"Teknium (@Teknium)","publisher":"X","date":"2026","claim_ids":[],"found_by":"","extra":{},"accessed_at":"2026-07-24T06:33:35.068Z","link_status":"ok","quote_status":"verified","prev":"0f40be50acfa089a453d486f7e51a2c84f59096d6733c0d56bbf2f212fb1b447","hash":"3b0ade8513feaf8cdb54867f2efeb676e8c422ce92ffc4efae01323485d85357"}]}