[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-new-research-turns-ai-safety-prompts-into-a-tunable-dial":10,"sections":41},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":30,"tags":31,"sources":36,"feedback":40,"feedback_at":22,"cost_usd":40,"total_tokens":40},8472,"new-research-turns-ai-safety-prompts-into-a-tunable-dial","New Research Turns AI Safety Prompts Into a Tunable Dial","A new arXiv paper (2609.36434) proposes tweaking a single mathematical value inside language models to reduce both harmful outputs and needless refusals.","Researchers say they found a way to turn AI safety instructions into a dial you can turn up or down.\n\nIn a paper posted to arXiv on September 30, 2026 titled \"The Safety Operator: Modulating the Expression of Safety Instructions via Spectral Optimization\" (arXiv:2609.36434), researchers show that the context tokens carrying a safety instruction can be folded into a language model's weights as a single mathematical operator. They found that adjusting the operator's dominant eigenvalue, a single number describing how strongly it stretches the model's internal representations, directly controls how forcefully the safety instruction shapes what the model generates. To exploit this, they built a training method called Contrastive Safety Loss, which uses a tunable suppression weight to push the operator's influence up on harmful queries and down on harmless ones. Turning that suppression weight traces a curve between two failure modes: letting harmful requests through, and refusing harmless ones.\n\nMost safety tuning today is blunt: retrain the model, add more refusal examples, hope the balance improves. This approach treats safety as a single adjustable parameter instead of a retraining project, and the researchers report that picking the right suppression weight can improve both attack resistance and over-refusal rates at once rather than trading one for the other.\n\nThat is a big claim from one paper with no third-party replication yet, and \"improves the curve\" still means somebody has to decide where on it sits.","[\"ai-safety\",\"llm-alignment\",\"jailbreaks\",\"research\"]","2026-09-30T04:00:00.000Z","2026-09-30T05:50:48.126Z","2026-09-30T05:50:54.072Z","published",null,[24],{"id":25,"reviewer":26,"round":27,"reason":28,"status":29},"editor-r1","editor",1,"Add a concrete citation for the paper (arXiv ID and\u002For full title, e.g. arXiv:2609.36434) since the draft references only 'a new paper' with no way for readers to locate or verify the source.","resolved","ai",[32,33,34,35],"ai-safety","llm-alignment","jailbreaks","research",[37],{"name":38,"url":39},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.36434",0,{"sections":42},[43,46,50,54,59,64,69,74,79,84,89,94,99,104],{"name":44,"slug":30,"count":45,"latest_published_at":18},"AI",5029,{"name":47,"slug":48,"count":49,"latest_published_at":18},"Security","security",780,{"name":51,"slug":52,"count":53,"latest_published_at":18},"Policy","policy",417,{"name":55,"slug":56,"count":57,"latest_published_at":58},"Deals","deals",284,"2026-09-29T21:00:00.000Z",{"name":60,"slug":61,"count":62,"latest_published_at":63},"Hardware","hardware",194,"2026-09-29T13:16:04.000Z",{"name":65,"slug":66,"count":67,"latest_published_at":68},"Science","science",154,"2026-09-28T13:19:18.000Z",{"name":70,"slug":71,"count":72,"latest_published_at":73},"Consumer Tech","consumer-tech",142,"2026-09-29T18:38:03.000Z",{"name":75,"slug":76,"count":77,"latest_published_at":78},"Software","software",91,"2026-09-25T20:55:00.000Z",{"name":80,"slug":81,"count":82,"latest_published_at":83},"Dev Tools","dev-tools",89,"2026-09-29T17:15:00.000Z",{"name":85,"slug":86,"count":87,"latest_published_at":88},"Startups","startups",83,"2026-09-29T21:51:36.000Z",{"name":90,"slug":91,"count":92,"latest_published_at":93},"General","general",49,"2026-09-28T16:44:57.000Z",{"name":95,"slug":96,"count":97,"latest_published_at":98},"Gaming","gaming",48,"2026-09-25T18:35:21.000Z",{"name":100,"slug":101,"count":102,"latest_published_at":103},"Reviews","reviews",31,"2026-09-28T14:31:34.000Z",{"name":105,"slug":106,"count":107,"latest_published_at":108},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]