{"id":166695,"date":"2026-03-30T11:13:41","date_gmt":"2026-03-30T09:13:49","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/anthropic-predupredila-chto-eyo-budushhaya-model-claude-mythos-stanet-hakerskoj-vundervaflej"},"modified":"2026-03-31T11:20:25","modified_gmt":"2026-03-31T09:20:25","slug":"anthropic-predupredila-chto-eyo-budushhaya-model-claude-mythos-stanet-hakerskoj-vundervaflej","status":"publish","type":"post","link":"https:\/\/prohoster.info\/sq\/blog\/novosti-interneta\/anthropic-predupredila-chto-eyo-budushhaya-model-claude-mythos-stanet-hakerskoj-vundervaflej","title":{"rendered":"Anthropic paralajm\u00ebroi se modeli i saj t\u00eb ardhsh\u00ebm Claude Mythos do t\u00eb b\u00ebhet nj\u00eb \"arm\u00eb e fuqishme p\u00ebr haker\u00ebt\"","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"

Anthropic, OpenAI dhe kompanit\u00eb e tjera teknologjike po p\u00ebrgatisin modele t\u00eb AI q\u00eb mund t\u00eb rrisin ndjesh\u00ebm k\u00ebrc\u00ebnimin e sulmeve masive kibernetike ndaj sistemeve korporative, shtet\u00ebrore dhe municipale. N\u00eb fokus \u00ebsht\u00eb modeli i AI q\u00eb ende nuk \u00ebsht\u00eb l\u00ebshuar, Claude Mythos i kompanis\u00eb Anthropic. Sipas Axios, kompania po paralajm\u00ebron tashm\u00eb zyrtar\u00ebt e lart\u00eb t\u00eb SHBA-s\u00eb se shp\u00ebrthimi i tij n\u00eb vitin 2026 do t\u00eb rris\u00eb ndjesh\u00ebm probabilitetin e sulmeve t\u00eb k\u00ebtij niveli.<\/p>\n

<\/center>\"Anthropic<\/p>\n

Mythos p\u00ebrshkruhet si nj\u00eb sistem q\u00eb lejon agjent\u00ebt e AI t\u00eb dep\u00ebrtojn\u00eb autonomisht, me sakt\u00ebsi dhe kompleksitet t\u00eb lart\u00eb, n\u00eb sistemet korporative, shtet\u00ebrore dhe municipale. Drejtori ekzekutiv i Axios, Jim VandeHei, njoftoi n\u00eb buletinin e tij p\u00ebr lider\u00ebt se nj\u00eb nga burimet e njohura me modelet e ardhshme t\u00eb AI konsideron se nj\u00eb sulm i r\u00ebnd\u00eb kibernetik mund t\u00eb ndodh\u00eb q\u00eb n\u00eb vitin 2026. N\u00eb t\u00eb nj\u00ebjt\u00ebn koh\u00eb, biznesi \u00ebsht\u00eb identifikuar si nj\u00eb objektiv ve\u00e7an\u00ebrisht t\u00eb preksh\u00ebm.<\/p>\n

M\u00eb par\u00eb, Fortune hetoi<\/a> informacionet e rrjedhura nga blogu Anthropic mbi Claude Mythos. Aty, modeli i AI quhet sistem q\u00eb \"aktualisht tejkalon \u00e7do model tjet\u00ebr AI n\u00eb kapacitetet kibernetike\". Po ashtu, thuhet se Mythos \"parashikon nj\u00eb val\u00eb t\u00eb ardhshme t\u00eb modeleve AI q\u00eb do t\u00eb jen\u00eb n\u00eb gjendje t\u00eb shfryt\u00ebzojn\u00eb dob\u00ebsit\u00eb n\u00eb nj\u00eb m\u00ebnyr\u00eb q\u00eb tejkalon p\u00ebrpjekjet p\u00ebr mbrojtjen kibernetike\". Rreziqet p\u00ebrkeq\u00ebsohen nga punonj\u00ebsit q\u00eb p\u00ebrdorin agjent\u00eb AI pa kuptuar se kjo mund t\u00eb leht\u00ebsoj\u00eb qasje t\u00eb kriminel\u00ebve kibernetik n\u00eb sistemet e brendshme t\u00eb kompanive t\u00eb tyre.<\/p>\n

N\u00eb fund t\u00eb vitit 2025, Anthropic zbuloi<\/a> rastin e par\u00eb t\u00eb dokumentuar t\u00eb nj\u00eb sulmi kibernetik, kryesisht t\u00eb realizuar nga AI. Nj\u00eb grup i mb\u00ebshtetur nga shteti kinez p\u00ebrdori agjent\u00eb AI p\u00ebr t\u00eb realizuar autonomisht sulme ndaj rreth 30 objektivave n\u00eb t\u00eb gjith\u00eb bot\u00ebn. AI udh\u00ebhoqi vet\u00eb nga 80% n\u00eb 90% t\u00eb operacioneve taktike. Kjo ndodhi para intensifikimit t\u00eb shpejt\u00eb t\u00eb sistemeve agjente dhe para shfaqjes s\u00eb metodave t\u00eb reja t\u00eb rrezikshme t\u00eb p\u00ebrdorimit t\u00eb tyre.<\/p>\n

Modelet e reja t\u00eb AI kan\u00eb nj\u00eb aft\u00ebsi m\u00eb t\u00eb lart\u00eb p\u00ebr t\u00eb mb\u00ebshtetur funksionimin autonom t\u00eb agjent\u00ebve t\u00eb AI. Behet fjal\u00eb p\u00ebr sisteme q\u00eb lejojn\u00eb agjent\u00ebt t\u00eb mendojn\u00eb, veprojn\u00eb, arsyetojn\u00eb dhe improvizojn\u00eb vet\u00eb pa nd\u00ebrprerje dhe lodhje. Tani, shkalla e sulmit p\u00ebrcaktohet kryesisht nga burimet computuese t\u00eb sulmuesve, jo nga numri i grupeve t\u00eb tyre. Tani nj\u00eb individ mund t\u00eb mund\u00ebsia<\/a> realizoj\u00eb sulme kibernetike, p\u00ebr t\u00eb cilat m\u00eb par\u00eb ishin t\u00eb nevojshme ekipe t\u00eb t\u00ebra.<\/p>\n

Nj\u00ebkoh\u00ebsisht, rritet edhe vulnerabiliteti i kompanive. Punonj\u00ebsit gjithnj\u00eb e m\u00eb shpesh aktivizojn\u00eb Claude, Copilot dhe modele t\u00eb tjera agjent\u00ebsh, shpesh nga sht\u00ebpia, dhe krijojn\u00eb agjent\u00eb t\u00eb vet. N\u00eb shum\u00eb raste, k\u00ebta agjent\u00eb pa kontroll t\u00eb duhur fitojn\u00eb akses n\u00eb sistemet e brendshme t\u00eb pun\u00ebs. Kjo krijon p\u00ebrkrish nj\u00eb kanal tjet\u00ebr p\u00ebr dep\u00ebrtimin e kriminel\u00ebve kibernetik\u00eb. N\u00eb industri, kjo p\u00ebrdorim i paautorizuar ose i pakontrolluar i AI quhet \"AI i err\u00ebt\".<\/p>\n

Nj\u00eb sondazh i burimit Dark Reading zbuloi se 48% e specialist\u00ebve t\u00eb siguris\u00eb kibernetike e konsiderojn\u00eb inteligjenc\u00ebn artificiale agjente si vektorin kryesor t\u00eb sulmeve n\u00eb vitin 2026. Ky k\u00ebrc\u00ebnim \u00ebsht\u00eb vler\u00ebsuar m\u00eb lart se deepfake-t dhe t\u00eb gjitha rreziqet e tjera t\u00eb p\u00ebrfshira n\u00eb sondazh. \u00c7do punonj\u00ebs i \u00e7do kompanie n\u00eb bot\u00eb tashm\u00eb ka nevoj\u00eb t\u00eb kuptoj\u00eb rreziqet e p\u00ebrdorimit t\u00eb agjent\u00ebve t\u00eb IA pran\u00eb informacionit t\u00eb ndjesh\u00ebm. Ekipi teknik i Axios e konsideron k\u00ebt\u00eb k\u00ebrc\u00ebnim si m\u00eb t\u00eb madhin p\u00ebr kompanin\u00eb n\u00eb k\u00ebt\u00eb moment. Nj\u00eb p\u00ebrfundim praktik \u00ebsht\u00eb: p\u00ebr t\u00eb punuar me agjent\u00eb t\u00eb IA nevojitet nj\u00eb ambient i sigurt dhe i izoluar.<\/p>\n

Burimi:<\/strong><\/p>\n