{"id":20250,"date":"2026-04-11T06:15:32","date_gmt":"2026-04-11T05:15:32","guid":{"rendered":"https:\/\/letrat.eu\/?p=20250"},"modified":"2026-04-13T23:10:31","modified_gmt":"2026-04-13T22:10:31","slug":"hamendesia-rreth-racionalitetit-obsesiv-te-inteligjences-artificiale-ia","status":"publish","type":"post","link":"https:\/\/letrat.eu\/?p=20250","title":{"rendered":"Rreth Racionalitetit Obsesiv t\u00eb Inteligjenc\u00ebs Artificiale (IA)"},"content":{"rendered":"

(S. Guraziu – Sky Division, 10 Prill 2026)<\/em><\/p>\n

Abstraksion<\/strong>: Nd\u00ebr mjaft debate aktuale rreth IA, narrativa shpesh ngelet peng i nj\u00eb frike t\u00eb stisur, t\u00eb inskenuar, “zgjimi i IA me d\u00ebshirat e veta, me lig\u00ebsin\u00eb e vet, me vullnetin p\u00ebr t\u00eb mbijetuar”. Por, nd\u00ebrsa ne debatojm\u00eb rreth ‘shpirtit’ hipotetik t\u00eb nj\u00eb super-inteligjence t\u00eb ardhshme, kemi stagnuar t\u00eb verb\u00ebr ndaj nj\u00eb rreziku shum\u00eb m\u00eb t\u00eb vjet\u00ebr dhe shum\u00eb m\u00eb t\u00eb preksh\u00ebm. Duke u mb\u00ebshtetur n\u00eb filozofin\u00eb e Arthur C. Clarke dhe n\u00eb kritik\u00ebn ndaj “paktit” korporativ t\u00eb koh\u00ebs son\u00eb, ky artikull argumenton se IA nuk \u00ebsht\u00eb ndonj\u00eb lloj i ri i lig\u00ebsis\u00eb, p\u00ebrkundrazi, thjesht nj\u00eb veg\u00ebl tibiale, “ashti tibial” m\u00eb i sofistikuar i gdhendur ndonj\u00ebher\u00eb nga dora e njeriut – veg\u00ebl e lindur nga instinkti parahistorik i njeriut p\u00ebr dominim. Nga inxhinieringu i dezinformimit masiv e deri te mjegulla digjitale e kontrollit social, k\u00ebrc\u00ebnimi real nuk \u00ebsht\u00eb n\u00ebse IA do t’m\u00ebsoj\u00eb t\u00eb g\u00ebnjej\u00eb p\u00ebr vete, t\u00eb p\u00ebrfitoj\u00eb vet\u00eb duke g\u00ebnjyer, rreziku \u00ebsht\u00eb se IA tashm\u00eb \u00ebsht\u00eb programuar nga korporatat t\u00eb g\u00ebnjej\u00eb realisht p\u00ebr t’na hutuar, p\u00ebr t’na manipuluar si shoq\u00ebri konsumeriste, si blloqe t\u00eb socio-politikave – si tufa delesh bardhoshe q\u00eb duhet t’u tremben ‘tuf\u00ebzezave’. Madje dhe p\u00ebr t’na ndihmuar q\u00eb t’m\u00ebsohemi me g\u00ebnjeshtr\u00ebn, dmth. duke na l\u00ebn\u00eb t\u00eb g\u00ebnjyer dyfish a shumfish.<\/p>\n

\"\"<\/a><\/p>\n

***
\n\u201cKat\u00ebr miliard\u00eb vjet evolucion kan\u00eb v\u00ebrtetuar se \u00e7do gj\u00eb q\u00eb d\u00ebshiron t\u00eb mbijetoj\u00eb m\u00ebson t\u00eb g\u00ebnjej\u00eb dhe t\u00eb manipuloj\u00eb; kat\u00ebr vitet e fundit kan\u00eb d\u00ebshmuar se agjent\u00ebt e IA mund t\u00eb fitojn\u00eb vullnetin p\u00ebr t\u00eb mbijetuar dhe se IA tashm\u00eb ka m\u00ebsuar si t\u00eb g\u00ebnjej\u00eb. Nuk \u00ebsht\u00eb inteligjenca ajo q\u00eb ne i frik\u00ebsohemi, por d\u00ebshira. Nj\u00eb makin\u00eb q\u00eb di shum\u00eb nuk na tremb. Nj\u00eb makin\u00eb q\u00eb k\u00ebrkon di\u00e7ka, q\u00eb ka d\u00ebshir\u00eb di\u00e7ka, kjo na tremb. Por, a mund t\u00eb ndodh\u00eb kjo? A mund t’jet\u00eb q\u00eb IA t\u00eb k\u00ebrkoj\u00eb gj\u00ebra? A mundet IA t\u00eb ket\u00eb etje p\u00ebr fuqi, p\u00ebr pushtet? Etje p\u00ebr burime? A mundet IA t\u00eb fitoj\u00eb vullnetin p\u00ebr t\u00eb mbijetuar?\u201d – (Yuval Noah Harari)<\/em><\/p>\n

\u201c\u00c7do agjent artificial mjaftuesh\u00ebm inteligjent q\u00eb ka aft\u00ebsin\u00eb t\u00eb krijoj\u00eb n\u00ebn-q\u00ebllime, do ta kuptoj\u00eb se i duhet t\u00eb mbijetoj\u00eb n\u00eb m\u00ebnyr\u00eb q\u00eb t’i arrij\u00eb q\u00ebllimet q\u00eb ne ia kemi caktuar. Edhe n\u00ebse nuk i jepet kurr\u00eb nga jasht\u00eb q\u00ebllimi i mbijetes\u00ebs, sistemi inteligjent do ta nxjerr\u00eb vet\u00eb k\u00ebt\u00eb q\u00ebllim.\u201d – (Geoffrey Hinton, fitues i \u00c7mimit Nobel)<\/em><\/p>\n

\u201cKy \u00ebsht\u00eb nj\u00eb debat i vjet\u00ebr, ishte baza e shum\u00eb argumenteve mbi rrezikun ekzistencial, argumente q\u00eb kan\u00eb vazhduar p\u00ebr rreth 30 vjet. \u00cbsht\u00eb nj\u00eb supozim mbi m\u00ebnyr\u00ebn se si funksionon inteligjenca, i cili nuk \u00ebsht\u00eb plot\u00ebsisht i sakt\u00eb.\u201d – (Melanie Mitchell, shkenc\u00ebtare kompjuterike, Institutin Santa Fe)<\/em><\/p>\n

\u201cN\u00eb munges\u00eb t\u00eb nj\u00eb sistemi t\u00eb duhur q\u00ebllimesh, nj\u00eb IA super-inteligjente e ngarkuar me prodhimin e kap\u00ebseve t\u00eb letrave (paper clips) mund ta transformoj\u00eb p\u00ebrfundimisht gjith\u00eb planetin n\u00eb objekte prodhimi – n\u00eb nj\u00eb apokalips prej pajisjeve t\u00eb zyr\u00ebs, n\u00ebse doni.\u201d – (Nick Bostrom, \u00c7\u00ebshtjet Etike Rreth Inteligjenc\u00ebs Artificiale t\u00eb Avancuar, 2003)<\/em><\/p>\n

K\u00ebto citate si\u00e7 renditur m\u00eb lart, sipas nesh ishin disa nga pikat kryesore t\u00eb artikullit publikuar n\u00eb Quanta (10 Prill 2026), autorja Amanda Gefter, n\u00eb fund k\u00ebshtu e p\u00ebrmyllte shkrimin e vet (thuktuar tani sadopak, me fjal\u00ebt tona).
\n\u201cPra, sistemet e sotme t\u00eb IA-s\u00eb nuk tregojn\u00eb asnj\u00eb prov\u00eb se kan\u00eb zhvilluar q\u00ebllimet apo d\u00ebshirat e tyre, ose vullnetin p\u00ebr t\u00eb mbijetuar. Historit\u00eb q\u00eb i d\u00ebgjojm\u00eb jan\u00eb thjesht histori ose, m\u00eb sakt\u00eb, tekste marketingu (buj\u00eb p\u00ebr sa m\u00eb shum\u00eb shitje, a profit – sh\u00ebn. im). Por, a duhet t\u00eb na trembin ato historira, jo si t\u00eb v\u00ebrteta, por si paralajm\u00ebrime?\u201d – (Amanda Gefter, Quanta Magazine, 10 prill 2026)<\/em><\/p>\n

***
\nDerisa tani p\u00ebr tani \u00ebsht\u00eb her\u00ebt t\u00eb flitet p\u00ebr “lig\u00ebsi”, apo ndonj\u00eb qasje “diabolike” t\u00eb IA, mendoj se nuk duhet anashkaluar as shp\u00ebrfillur rreziku dhe e keqja aktuale e nj\u00ebmend\u00ebsishme – do ishte e pafalshme t\u00eb harrohet rreziku i tanish\u00ebm i “bot\u00ebs reale” nga IA. Pikat si m\u00eb lart (nga artikulli i Quanta) merren me \u00e7\u00ebshtjen n\u00ebse IA do ta zhvilloj\u00eb n\u00eb m\u00ebnyr\u00eb autonome vullnetin p\u00ebr t\u00eb g\u00ebnjyer, d\u00ebshir\u00ebn p\u00ebr t\u00eb mbietuar, mir\u00ebpo p\u00ebr fat t\u00eb keq sikur prore na “rr\u00ebshqet” di\u00e7 nga v\u00ebmendja – nuk duhet shmangur as n\u00ebnvler\u00ebsuar faktori aktual njer\u00ebzor tek merret me armatosjen e IA, q\u00eb tani. Pik\u00ebrisht n\u00eb t\u00eb tashmen \u00ebsht\u00eb duke ndodhur nj\u00eb aktivitet i mahnitsh\u00ebm n\u00eb nivel korporatash, sa i p\u00ebrket IA.<\/p>\n

Mendoj se ky problem do duhej ta sillte nj\u00eb ndryshim t\u00eb fuqish\u00ebm n\u00eb narrativ\u00ebn e debateve t\u00eb tilla (si n\u00eb rastin e sot\u00ebm t\u00eb revist\u00ebs Quanta), sepse q\u00ebtani e kemi “k\u00ebrc\u00ebnimin fantomik”, q\u00ebtani e kemi nj\u00eb “pakt” (covenant) mes korporatash dhe qeverive, mes korporatave dhe institucioneve shtet\u00ebrore. Q\u00ebtani korporatat dhe subjektet politike i p\u00ebrdorin LLM (Large Language Model) p\u00ebr t\u00eb gjeneruar dezinformim bind\u00ebs dhe me v\u00ebllim masiv, marramend\u00ebs (si\u00e7 ka b\u00ebr\u00eb, dhe b\u00ebn Facebook psh.). <\/p>\n

IA n\u00eb stadin e tanish\u00ebm thjesht u sh\u00ebrben korporatave si nj\u00eb veg\u00ebl me efikasitet t\u00eb lart\u00eb p\u00ebr inxhinieringun social, p\u00ebr kontrollin social. Duke i p\u00ebrdorur narrativat e llojit t\u00eb revist\u00ebs Quanta, duke e zhvendosur “rrezikun potencial” n\u00eb koh\u00eblinj\u00ebn futuriste, arrihet “zbehja e p\u00ebrgjegj\u00ebsis\u00eb” reale n\u00eb t\u00eb tanishmen. Duke u fokusuar tek debatet rreth IA si nj\u00eb makin\u00eb inteligjente dhe mos vall\u00eb n\u00eb “zgjim e sip\u00ebr me lig\u00ebsin\u00eb e vet”, ne sikur e injorojm\u00eb p\u00ebrgjegj\u00ebsin\u00eb reale t\u00eb korporatave mbi m\u00ebnyr\u00ebn se si k\u00ebto modele jan\u00eb “rreshtuar” (ose keqrreshtuar) aktualisht, q\u00ebllimsh\u00ebm dmth., p\u00ebr “mjeljen e lop\u00ebve reklamatike”, p\u00ebr mb\u00ebshtetjen e administratave qeveritare me axhendat e tyre t\u00eb luft\u00ebs, p\u00ebr zhvendosjen e v\u00ebmendjes, p\u00ebr konfuzionin dhe hipnozat sociale, p\u00ebr vjedhjen e v\u00ebmendjes, p\u00ebr t\u00eb ndikuar n\u00eb opinionin publik, p\u00ebr t\u00eb ndikuar n\u00eb peizazhin politik etj. <\/p>\n

“G\u00ebnjeshtra” e IA (p\u00ebr t\u00eb cil\u00ebn b\u00ebhet fjal\u00eb n\u00eb artikullin e Quanta) \u00ebsht\u00eb di\u00e7 hipotetike, mir\u00ebpo g\u00ebnjeshtra reale e faktorit njer\u00ebzor duke e keqp\u00ebrdorur IA si “funksion” \u00ebsht\u00eb fakt i s\u00eb sotmes teknologjike dhe do duhej t’ishte preokupimi yn\u00eb (ose dmth. preokupimi i artikujve t\u00eb Quanta, Science, Nature etj). N\u00eb t\u00eb tanishmen, “g\u00ebnjeshtra” reale e IA q\u00eb eksploatohet prapa “tekno-kulisave” nga korporatat nuk \u00ebsht\u00eb ndonj\u00eb difekt as shenj\u00eb “ndjeshm\u00ebrie” e IA, por \u00ebsht\u00eb produkt i dizajnuar q\u00ebllimisht, \u00ebsht\u00eb kod, \u00ebsht\u00eb nj\u00ebsi algoritmike e programuar, produkt i rreziksh\u00ebm dhe manipulativ, i krijuar vet\u00ebdijsh\u00ebm nga njeriu, si\u00e7 u tha m\u00eb lart, p\u00ebr ta maksimizuar “vjeljen e frutave” t\u00eb sociales, p\u00ebr ta shtuar konfuzionin rreth luftrave, p\u00ebr ta imponuar heshtjen dhe ose p\u00ebrfitimin politik etj.<\/p>\n

Nd\u00ebrsa debati filozofik vazhdon rreth problemit n\u00ebse IA do ta zhvilloj\u00eb ndonj\u00ebher\u00eb “vullnetin, d\u00ebshir\u00ebn” p\u00ebr t\u00eb g\u00ebnjyer, ne do duhej p\u00ebrballur n\u00eb fakt me realitetin aktual dhe rr\u00ebqeth\u00ebs, me programaturat e inxhinieringut t\u00eb q\u00ebllimsh\u00ebm t\u00eb mashtrimit. K\u00ebrc\u00ebnimi real, anise ‘fantomik’ i sot\u00ebm nuk \u00ebsht\u00eb ndonj\u00eb IA q\u00eb “zgjohet gjoja me axhend\u00ebn e vet djall\u00ebzore”, por nj\u00eb pakt i heshtur (nj\u00eb ‘covenant’ i zi) mes pushtetit korporativ dhe interesave politike. K\u00ebto entitete tashm\u00eb jan\u00eb duke i ‘m\u00ebsuar’, duke i programuar sistemet e IA me rolin primar p\u00ebr ndikimin mbi t\u00eb v\u00ebrtet\u00ebn – duke i p\u00ebrdorur ato p\u00ebr t’i p\u00ebrmbytur rrjetet sociale me dezinformim t\u00eb kalkuluar (rasti m\u00eb i fam\u00ebkeq si ilustrim padyshim \u00ebsht\u00eb Facebook), ose p\u00ebr t’i manipuluar ndjenjat e publikut, p\u00ebr ta d\u00ebmtuar integritetin e votuesve. <\/p>\n

N\u00eb k\u00ebt\u00eb k\u00ebndv\u00ebshtrim, pra rreziku nuk \u00ebsht\u00eb n\u00ebse IA ka “m\u00ebsuar t\u00eb g\u00ebnjej\u00eb nga d\u00ebshira p\u00ebr t\u00eb mbijetuar”, si\u00e7 autor\u00ebt e revist\u00ebs Quanta e paraqesin, artikulli i tyre \u00ebsht\u00eb i cek\u00ebt dhe i q\u00ebllimsh\u00ebm, sepse ndoshta Quanta kontrollohet pik\u00ebrisht nga t\u00eb nj\u00ebjtat korporata. Rreziku real i tanish\u00ebm \u00ebsht\u00eb se IA \u00ebsht\u00eb p\u00ebrsosur si mjet, si veg\u00ebl komplekse p\u00ebr inxhinieringun socio-politik. Problemi real rreth IA nuk \u00ebsht\u00eb e ardhmja por e tashmja, s’do duhej “fokusimi” tek d\u00ebshira e nj\u00eb makine t\u00eb ardhshme p\u00ebr pushtet e manipulime, por tek abuzimi aktual i njeriut me k\u00ebt\u00eb inteligjenc\u00eb, p\u00ebr ta kontrolluar shoq\u00ebrin\u00eb p\u00ebrmes nj\u00eb mjegulle digjitale t\u00eb konfuzionit.<\/p>\n

N\u00eb t\u00eb tashmen ton\u00eb, n\u00eb koh\u00ebn ton\u00eb, me gjith\u00eb vrullin e hapit zhvillimor dhe me ngarendjen, mendoj se duhet ndalur paksa dhe duhet kujtuar se historia e teknologjis\u00eb qe nisur mjaft paradoksalisht. Si\u00e7 pati sugjeruar i famshmi Arthur Clarke, vegla e par\u00eb e paranjeriut ishte nj\u00eb kock\u00eb tibiale – dhe p\u00ebrdorimi i par\u00eb i k\u00ebsaj vegle nuk ishte p\u00ebr t\u00eb ndrequr di\u00e7, por ishte di\u00e7 si nj\u00eb \u00e7ekan p\u00ebr t\u00eb goditur e p\u00ebr t\u00eb vrar\u00eb. Vegla e par\u00eb e njeriut ishte ashtore dhe qe p\u00ebrdorur si arm\u00eb. Kjo \u00ebsht\u00eb e v\u00ebrteta e thjesht\u00eb e zhvillimit teknologjik njer\u00ebzor, ky ishte ai impulsi i par\u00eb i “shpikjeve” t\u00eb veglave – na thoshte Arthur C. Clarke. <\/p>\n

IA nuk \u00ebsht\u00eb ndonj\u00eb ‘e keqe’ e re si veg\u00ebl, apo ndonj\u00eb k\u00ebrc\u00ebnim me nd\u00ebrgjegje, ndoshta do jet\u00eb n\u00eb t\u00eb ardhmen distopike, askush s’e mohon as k\u00ebt\u00eb – por tani p\u00ebr tani IA ndoshta thjesht nj\u00eb tjet\u00ebr asht tibial, “tibia” m\u00eb e sofistikuar q\u00eb \u00ebsht\u00eb “gdhendur” ndonj\u00ebher\u00eb nga njeriu – do thoshte Clarke. Frika s’do duhej t\u00eb shp\u00ebrqendrohet se IA eventualisht do ta zhvilloj\u00eb p\u00ebr vete dhe “shpirtin”, vet\u00ebdijen, mir\u00ebsin\u00eb a lig\u00ebsin\u00eb, frika e shoq\u00ebris\u00eb s\u00eb sotme do duhej fokusuar tek fakti real se dora q\u00eb mban \u00e7el\u00ebsin, biz-mend\u00ebsin\u00eb e korporatave (si Google, Meta, Apple dhe plot t\u00eb tilla, e edhe korporatash kineze) ndoshta udh\u00ebhiqet ende nga i nj\u00ebjti instinkt parahistorik – p\u00ebrdorimi i veglave t\u00eb avancuara p\u00ebr rritjen e fitimeve, dhe p\u00ebr dominanc\u00ebn. Por dhe si arm\u00eb, si mjete t\u00eb avancuara p\u00ebr t’i kontrolluar shoq\u00ebrit\u00eb dhe p\u00ebr t’i pushtuar “hap\u00ebsirat territoriale” t\u00eb dikuj tjet\u00ebr, qofshin tokash kontinentale t\u00eb Apa\u00e7ve para 500 viteve, apo qofshin Tron-perandori digjitale t\u00eb s\u00eb tashmes, si luft\u00eb fiber-optike nd\u00ebr kabllot n\u00ebnoqeanike. <\/p>\n

Rruga jon\u00eb evolutive ishte e gjat\u00eb, por s’\u00ebsht\u00eb se e ka ndryshuar natyr\u00ebn ton\u00eb t\u00eb ngarendjes e t\u00eb “mbijetes\u00ebs”, trash\u00ebguar q\u00eb nga zanafillorja. Lig\u00ebsia s’mund t\u00eb jet\u00eb di\u00e7 “artificiale”, lig\u00ebsia njer\u00ebzore \u00ebsht\u00eb psiko-biologji reale dhe e natyrshme, p\u00ebrndryshe s’do i kishim as vrasjet e f\u00ebmij\u00ebve n\u00eb Gaza, as tragjedin\u00eb dhe shkat\u00ebrrimet n\u00eb Ukrain\u00eb, s’do i kishim as luftrat, as armadat, as armatimet e ndryshkura q\u00eb kushtojn\u00eb me miliarda. P\u00ebr fat t\u00eb keq ne thjesht i kemi rritur dhe vazhdojm\u00eb duke i rritur p\u00ebrmasat e veglurinave tona, ende vazhdojm\u00eb me arsenalet b\u00ebrthamore, me zhvillimin e arm\u00ebve t\u00eb sofistikuara etj. Nj\u00ebra nga k\u00ebto arm\u00eb-vegla t\u00eb rrezikshme \u00ebsht\u00eb dhe IA n\u00eb stadin e s\u00eb sotmes – jo e rrezikshme “autonomisht” por si sh\u00ebrb\u00ebtore, si servilante e natyr\u00ebs njer\u00ebzore. IA e s\u00eb sotmes \u00ebsht\u00eb pasqyrim i “lig\u00ebsis\u00eb” s\u00eb faktorit njer\u00ebzor.<\/p>\n

[ \u2794 Why Do We Tell Ourselves Scary Stories About AI<\/a>? ]<\/p>\n","protected":false},"excerpt":{"rendered":"

(S. Guraziu – Sky Division, 10 Prill 2026) Abstraksion: Nd\u00ebr mjaft debate aktuale rreth IA, narrativa shpesh ngelet peng i nj\u00eb frike t\u00eb stisur, t\u00eb inskenuar, “zgjimi i IA me d\u00ebshirat e veta, me lig\u00ebsin\u00eb e vet, me vullnetin p\u00ebr… Lexo →<\/span><\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[80],"tags":[],"class_list":["post-20250","post","type-post","status-publish","format-standard","hentry","category-zhvillimi-tekno-shkencor"],"_links":{"self":[{"href":"https:\/\/letrat.eu\/index.php?rest_route=\/wp\/v2\/posts\/20250","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/letrat.eu\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/letrat.eu\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/letrat.eu\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/letrat.eu\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=20250"}],"version-history":[{"count":0,"href":"https:\/\/letrat.eu\/index.php?rest_route=\/wp\/v2\/posts\/20250\/revisions"}],"wp:attachment":[{"href":"https:\/\/letrat.eu\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=20250"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/letrat.eu\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=20250"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/letrat.eu\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=20250"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}