{"id":118040,"date":"2026-09-22T13:57:21","date_gmt":"2026-09-22T10:57:21","guid":{"rendered":"https:\/\/prizrenpost.com\/al\/a-mund-ta-zhduke-inteligjenca-artificiale-njerezimin\/"},"modified":"2026-09-22T13:57:21","modified_gmt":"2026-09-22T10:57:21","slug":"a-mund-ta-zhduke-inteligjenca-artificiale-njerezimin","status":"publish","type":"post","link":"https:\/\/prizrenpost.com\/al\/a-mund-ta-zhduke-inteligjenca-artificiale-njerezimin\/","title":{"rendered":"A mund ta zhduk\u00eb inteligjenca artificiale njer\u00ebzimin?"},"content":{"rendered":"
\"A<\/figure>\n

Debati n\u00ebse inteligjenca artificiale mund t\u00eb k\u00ebrc\u00ebnoj\u00eb vet\u00eb mbijetes\u00ebn e njer\u00ebzimit \u00ebsht\u00eb rindezur dit\u00ebt e fundit, pas paralajm\u00ebrimeve publike nga profesionist\u00ebt e IA-s\u00eb se zhvillimi i sistemeve gjithnj\u00eb e m\u00eb t\u00eb fuqishme po na \u00e7on rreziksh\u00ebm pran\u00eb shkat\u00ebrrimit.<\/p>\n

Arsyeja p\u00ebr k\u00ebt\u00eb ishte vendimi i Jacob Coxon, nj\u00eb punonj\u00ebs i kompanis\u00eb s\u00eb inteligjenc\u00ebs artificiale Anthropic, p\u00ebr t\u2019u larguar nga kompania, duke argumentuar se drejtimi i hulumtimit ishte shum\u00eb i rreziksh\u00ebm. Pak m\u00eb von\u00eb, kolegu i tij Evan Humpinger vler\u00ebsoi se probabiliteti q\u00eb nj\u00eb IA e p\u00ebrparuar t\u00eb \u00e7oj\u00eb n\u00eb zhdukjen e specieve njer\u00ebzore brenda dekad\u00ebs s\u00eb ardhshme tejkalon 10%.<\/p>\n

Si\u00e7 raporton CNN n\u00eb analiz\u00ebn e saj, k\u00ebto pretendime kan\u00eb provokuar reagime t\u00eb forta nga politikan\u00ebt, qarqet e teknologjis\u00eb dhe figurat publike, duke nxjerr\u00eb n\u00eb pah nj\u00eb pyetje themelore: n\u00ebse inteligjenca artificiale mund t\u00eb zhduk\u00eb v\u00ebrtet njer\u00ebzimin, si do t\u00eb ndodhte sakt\u00ebsisht kjo?<\/p>\n

\u201cEkspert\u00ebt\u201d m\u00eb pesimist\u00eb flasin p\u00ebr skenar\u00eb q\u00eb p\u00ebrfshijn\u00eb arm\u00eb biologjike, robot\u00eb autonom\u00eb ose marrjen e kontrollit mbi infrastruktur\u00ebn kritike. Shkenc\u00ebtar\u00eb t\u00eb tjer\u00eb, megjithat\u00eb, argumentojn\u00eb se k\u00ebto parashikime bazohen n\u00eb supozime rreth teknologjive q\u00eb ende nuk ekzistojn\u00eb dhe se nuk ka faktor\u00eb specifik\u00eb q\u00eb do ta b\u00ebnin t\u00eb mundur nj\u00eb sekuenc\u00eb t\u00eb till\u00eb ngjarjesh.<\/p>\n

Nj\u00eb nga skenar\u00ebt m\u00eb t\u00eb diskutuar ka t\u00eb b\u00ebj\u00eb me biologjin\u00eb. Disa studiues kan\u00eb sugjeruar q\u00eb inteligjenca artificiale shum\u00eb e p\u00ebrparuar mund t\u00eb ndihmoj\u00eb n\u00eb krijimin e nj\u00eb patogjeni t\u00eb rreziksh\u00ebm ose t\u00eb p\u00ebrdor\u00eb njer\u00ebzit si \u201cnd\u00ebrmjet\u00ebs\u201d p\u00ebr zhvillimin e arm\u00ebve biologjike.<\/p>\n

Thomas Larsen, nj\u00eb studiues n\u00eb Projektin AI Futures dhe nj\u00eb ish-an\u00ebtar i Institutit t\u00eb K\u00ebrkimit t\u00eb Inteligjenc\u00ebs Makinerie, argumenton se nj\u00eb sistem shum\u00eb m\u00eb i sofistikuar se modelet e sotme mund t\u00eb manipuloj\u00eb njer\u00ebzit q\u00eb t\u00eb kryejn\u00eb eksperimente n\u00eb em\u00ebr t\u00eb tij. Ky shqet\u00ebsim p\u00ebrkeq\u00ebsohet nga fakti se studiuesit tashm\u00eb po p\u00ebrdorin modele t\u00eb IA-s\u00eb p\u00ebr pun\u00eb shkencore, duke p\u00ebrfshir\u00eb fushat q\u00eb lidhen me biologjin\u00eb dhe k\u00ebrkimin farmaceutik.<\/p>\n

Megjithat\u00eb, shum\u00eb shkenc\u00ebtar\u00eb argumentojn\u00eb se kalimi nga njohurit\u00eb teorike n\u00eb nj\u00eb arm\u00eb biologjike reale, operacionale dhe shum\u00eb t\u00eb rrezikshme \u00ebsht\u00eb nj\u00eb hap i madh. Eric Singh, nj\u00eb profesor i t\u00eb m\u00ebsuarit automatik n\u00eb Universitetin Carnegie Mellon, i cili mban nj\u00eb doktoratur\u00eb si n\u00eb shkenc\u00ebn kompjuterike ashtu edhe n\u00eb mikrobiologji, e krahason procesin me p\u00ebrpjekjen p\u00ebr t\u00eb nd\u00ebrtuar Lego pa udh\u00ebzime.<\/p>\n

\u201cNuk \u00ebsht\u00eb sikur t\u2019i hedh\u00ebsh t\u00eb gjitha pjes\u00ebt n\u00eb tok\u00eb dhe ato thjesht bashkohen,\u201d shpjegon ai, duke v\u00ebn\u00eb n\u00eb dukje se biologjia k\u00ebrkon sakt\u00ebsi ekstreme, kushte t\u00eb ve\u00e7anta dhe procese komplekse. N\u00eb t\u00eb nj\u00ebjt\u00ebn koh\u00eb, ai v\u00ebren se ekzistojn\u00eb tashm\u00eb mekanizma t\u00eb rrept\u00eb kontrolli rreth laborator\u00ebve, materialeve dhe zinxhir\u00ebve t\u00eb furnizimit, t\u00eb cil\u00ebt veprojn\u00eb si pengesa t\u00eb r\u00ebnd\u00ebsishme p\u00ebr zhvillimin e k\u00ebrc\u00ebnimeve t\u00eb tilla.<\/p>\n

Nj\u00eb skenar i dyt\u00eb i konsideruar nga mb\u00ebshtet\u00ebsit e teoris\u00eb s\u00eb \u201ck\u00ebrc\u00ebnimit ekzistencial\u201d ka t\u00eb b\u00ebj\u00eb me krijimin e robot\u00ebve autonom\u00eb q\u00eb mund t\u00eb riprodhohen dhe t\u00eb veprojn\u00eb pa nd\u00ebrhyrjen njer\u00ebzore.<\/p>\n

Nate Soares, president i Institutit t\u00eb K\u00ebrkimit t\u00eb Inteligjenc\u00ebs s\u00eb Makinerive, argumenton se pika e kthes\u00ebs do t\u00eb ishte momenti kur nj\u00eb IA mund t\u00eb krijonte infrastruktur\u00ebn q\u00eb i nevojitet vet\u00eb: fabrika, sisteme energjie dhe makina t\u00eb reja. \u201cKur keni robot\u00eb q\u00eb mund t\u00eb nd\u00ebrtojn\u00eb infrastruktur\u00ebn p\u00ebr t\u00eb prodhuar m\u00eb shum\u00eb robot\u00eb, at\u00ebher\u00eb po flasim p\u00ebr nj\u00eb form\u00eb t\u00eb re mekanike t\u00eb jet\u00ebs\u201d, argumenton ai.<\/p>\n

Megjithat\u00eb, ky skenar \u00ebsht\u00eb larg realitetit t\u00eb teknologjis\u00eb s\u00eb sotme. Pavar\u00ebsisht investimeve t\u00eb m\u00ebdha t\u00eb manjat\u00ebve si Elon Musk n\u00eb robot\u00eb humanoide, k\u00ebto sisteme jan\u00eb ende subjekt i kufizimeve serioze dhe nuk kan\u00eb aft\u00ebsin\u00eb p\u00ebr t\u00eb prodhuar dhe zgjeruar n\u00eb m\u00ebnyr\u00eb autonome.<\/p>\n

A mund t\u00eb kontrolloj\u00eb IA arm\u00ebt b\u00ebrthamore?<\/p>\n

Nj\u00eb pyetje tjet\u00ebr ka t\u00eb b\u00ebj\u00eb me arm\u00ebt b\u00ebrthamore dhe n\u00ebse nj\u00eb inteligjenc\u00eb artificiale e p\u00ebrparuar mund t\u00eb fitoj\u00eb qasje n\u00eb to.<\/p>\n

Heidi Klaaf, shkenc\u00ebtare kryesore n\u00eb Institutin AI Now dhe ish-inxhiniere sigurie n\u00eb OpenAI, thekson se sistemet b\u00ebrthamore jan\u00eb projektuar me protokolle sigurie jasht\u00ebzakonisht t\u00eb rrepta dhe jan\u00eb kryesisht t\u00eb izoluara nga interneti. Edhe virusi kibernetik Stuxnet, i cili d\u00ebmtoi objektet b\u00ebrthamore t\u00eb Iranit, k\u00ebrkoi qasje fizike n\u00eb sistem n\u00ebp\u00ebrmjet nj\u00eb USB-je, thot\u00eb ai.<\/p>\n

Herbert Lin, nj\u00eb studiues dhe ekspert sigurie n\u00eb Universitetin e Stanfordit, argumenton se k\u00ebrc\u00ebnimi i v\u00ebrtet\u00eb q\u00ebndron ende te vet\u00eb arm\u00ebt b\u00ebrthamore dhe vendimet njer\u00ebzore rreth tyre, jo te nj\u00eb skenar hipotetik i IA-s\u00eb q\u00eb merr kontrollin e sistemeve.<\/p>\n

P\u00ebr disa studiues, k\u00ebrc\u00ebnimi i v\u00ebrtet\u00eb nuk \u00ebsht\u00eb domosdoshm\u00ebrisht inteligjenca artificiale e sotme, por mund\u00ebsia e nj\u00eb t\u00eb ashtuquajture \u201csuperinteligjenc\u00eb artificiale\u201d. Teoria thot\u00eb se n\u00ebse nj\u00eb sistem fiton aft\u00ebsin\u00eb p\u00ebr t\u00eb p\u00ebrmir\u00ebsuar aft\u00ebsit\u00eb e veta vet\u00eb, evolucioni i tij mund t\u00eb b\u00ebhet aq i shpejt\u00eb sa njer\u00ebzit nuk mund ta kontrollojn\u00eb m\u00eb at\u00eb.<\/p>\n

Soares argumenton se problemi nuk \u00ebsht\u00eb nj\u00eb IA q\u00eb ka q\u00ebllime \u201carmiq\u00ebsore\u201d, por nj\u00eb sistem q\u00eb ndjek nj\u00eb q\u00ebllim pa kuptuar vler\u00ebn njer\u00ebzore. \u201cN\u00ebse nj\u00eb sistem d\u00ebshiron m\u00eb shum\u00eb burime kompjuterike p\u00ebr t\u00eb arritur q\u00ebllimin e tij dhe nuk ka arsye p\u00ebr t\u00eb mbrojtur njer\u00ebzit, at\u00ebher\u00eb ai mund ta shnd\u00ebrroj\u00eb bot\u00ebn n\u00eb di\u00e7ka n\u00eb t\u00eb cil\u00ebn nuk do t\u00eb jemi n\u00eb gjendje t\u00eb mbijetojm\u00eb\u201d, argumenton ai.<\/p>\n

Mosmarr\u00ebveshja mbi t\u00eb ardhmen e inteligjenc\u00ebs artificiale nuk ka t\u00eb b\u00ebj\u00eb vet\u00ebm me teknologjin\u00eb, por edhe me m\u00ebnyr\u00ebn se si njer\u00ebzit e perceptojn\u00eb fuqin\u00eb dhe kufijt\u00eb e saj. Thomas Larsen u p\u00ebrpoq t\u2019u jepte nj\u00eb form\u00eb m\u00eb konkrete skenar\u00ebve t\u00eb mundsh\u00ebm t\u00eb fundit t\u00eb bot\u00ebs p\u00ebrmes analizave t\u00eb tij t\u00eb \u201cIA 2027\u201d dhe \u201cIA 2040\u201d, duke shqyrtuar rrug\u00eb t\u00eb ndryshme t\u00eb zhvillimit dhe rregullimit t\u00eb inteligjenc\u00ebs artificiale. Megjithat\u00eb, edhe n\u00eb skenarin e tij m\u00eb optimist (ku ka nj\u00eb marr\u00ebveshje globale se si duhet t\u00eb zhvillohet IA dhe njer\u00ebzimi ka filluar tashm\u00eb t\u00eb zgjerohet n\u00eb hap\u00ebsir\u00eb) rezultati mbetet, sipas tij, dominimi i sistemeve mekanike n\u00eb nj\u00eb pik\u00eb t\u00eb paspecifikuar n\u00eb t\u00eb ardhmen.<\/p>\n

Ai deklaron se nuk ka dyshim se rruga drejt superinteligjenc\u00ebs do t\u00eb vazhdoj\u00eb, p\u00ebrve\u00e7 n\u00ebse merren vendime t\u00eb vet\u00ebdijshme p\u00ebr t\u00eb ndryshuar drejtimin e zhvillimit.<\/p>\n

Nga ana tjet\u00ebr, Herbert Lin beson se nj\u00eb pjes\u00eb e madhe e distanc\u00ebs midis mb\u00ebshtet\u00ebsve t\u00eb skenar\u00ebve t\u00eb fundit t\u00eb bot\u00ebs dhe skeptik\u00ebve lind nga m\u00ebnyrat e ndryshme n\u00eb t\u00eb cilat njer\u00ebzit e p\u00ebrjetojn\u00eb zhvillimin teknologjik. Si\u00e7 shpjegon ai, ka di\u00e7ka ve\u00e7an\u00ebrisht emocionuese n\u00eb lidhje me p\u00ebrvoj\u00ebn e krijimit t\u00eb nj\u00eb programi q\u00eb duket se \u201cgjall\u00ebrohet\u201d dhe i p\u00ebrgjigjet komandave t\u00eb dikujt. \u00cbsht\u00eb pik\u00ebrisht kjo ndjenj\u00eb e zbulimit t\u00eb papar\u00eb, sipas Lin, q\u00eb mund t\u2019i \u00e7oj\u00eb disa studiues t\u00eb projektojn\u00eb p\u00ebrparimin mbres\u00ebl\u00ebn\u00ebs t\u00eb sot\u00ebm n\u00eb inteligjenc\u00ebn artificiale n\u00eb nj\u00eb t\u00eb ardhme pa kufij, madje edhe n\u00eb skenar\u00eb ekstrem\u00eb ku zhvillimi teknologjik i shp\u00ebton \u00e7do kontrolli njer\u00ebzor. \/tesheshi.com\/<\/p>\n

\n

Burimi: tesheshi<\/strong><\/p>\n","protected":false},"excerpt":{"rendered":"

Debati n\u00ebse inteligjenca artificiale mund t\u00eb k\u00ebrc\u00ebnoj\u00eb vet\u00eb mbijetes\u00ebn e njer\u00ebzimit \u00ebsht\u00eb rindezur dit\u00ebt e fundit, pas paralajm\u00ebrimeve publike nga profesionist\u00ebt e IA-s\u00eb se zhvillimi i sistemeve gjithnj\u00eb e m\u00eb t\u00eb fuqishme po na \u00e7on rreziksh\u00ebm pran\u00eb shkat\u00ebrrimit. Arsyeja p\u00ebr k\u00ebt\u00eb ishte vendimi i Jacob Coxon, nj\u00eb punonj\u00ebs i kompanis\u00eb s\u00eb inteligjenc\u00ebs artificiale Anthropic, p\u00ebr […]<\/p>\n","protected":false},"author":1,"featured_media":109453,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6],"tags":[],"class_list":["post-118040","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-teknologji"],"_links":{"self":[{"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/posts\/118040","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/comments?post=118040"}],"version-history":[{"count":0,"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/posts\/118040\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/media\/109453"}],"wp:attachment":[{"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/media?parent=118040"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/categories?post=118040"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/prizrenpost.com\/al\/wp-json\/wp\/v2\/tags?post=118040"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}