Umdali wokuqukethwe kwesayensi owaziwayo phesheya uHank Green uthe uzonciphisa ijubane lokudala okuqukethwe ngemuva kokugxekwa abalandeli ngenxa yokusebenzisa i-AI.
Ngokwe-IT之家, ngaphandle kwesiteshi sakhe, uGreen uphinde aphathe iziteshi ezidumile zesayensi ku-YouTube i-SciShow ne-Crash Course. I-SciShow yakhiwe iziteshi eziningi ezigxile emikhakheni ehlukene yesayensi, kanti i-Crash Course ichaza ngokucacile nangokujulile ulwazi oluhlukahlukene, okuhlanganisa izincwadi zaseLatin America nezibalo zokufunda.
Ngaphezu kwalokho, yena nomfowabo uJohn Green basungula ndawonye isiteshi i-Vlogbrothers.
UGreen ushicilele umbhalo omude ku-Reddit ngoLwesihlanu ngesikhathi sendawo, ethi imvamisa yokubuyekeza kwakhe ku-YouTube izokwehla kakhulu esikhathini esizayo.
Ubhale wathi: “Abantu bangalindela ukuthi izibuyekezo ze-Hankschannel zinciphe, futhi kungenzeka ime isikhashana. Ngokuzayo ingase ibuye, isekelwe ekubhaleni nasecwaningweni, noma imvamisa yokubuyekeza ivele yehle kakhulu.”
Uphinde wathi imidlalo yakhe emibili yansuku zonke yamagama aku-inthanethi, i-SMUSH ne-4x3, nayo izomisa izibuyekezo.
Le mpikiswano ngokusebenzisa kukaGreen i-AI iqale ekuqaleni kwalolo suku. Ngaleso sikhathi, wadalula enkundleni yezokuxhumana i-X ukuthi wake wasebenzisa i-ChatGPT ye-OpenAI ukwenza ucwaningo lwemininingwane yeskripthi sevidiyo.
Lokhu kwethusa abasebenzisi abaningi be-X, abanye bethi uGreen wonakalise ukwethembeka kwakhe ngokudala okuqukethwe ngaphandle kokudalulela izibukeli ukuthi usebenzise i-AI.
Njengoba impikiswano iqhubeka ikhula, uGreen wasusa lokho okuthunyelwe wabe eseya ku-Reddit eyophendula ngalolu daba. Uthe “uzizwa enecala elikhulu kakhulu” ngoba “uphoxe abantu abaningi”.
Uthe: “Ngempela ngithembele kakhulu ku-AI njengethuluzi lokusiza ocwaningweni. Iwusizo kakhulu kulokhu, ngoba ingisiza ngithole ngokushesha amaphepha amaningi engangingawazi ngaphambilini. Kodwa ngicabanga ukuthi lokhu kwalimaza ukudala kwami, ngoba kwangiphuca inqubo yokuhlola nokuqonda isihloko ngendlela yami.”
Njengoba izinkampani zobuchwepheshe ziqhubeka nokukhuthaza ukufakwa kwe-AI empilweni yabantu yansuku zonke, abagxeki abaningi baya ngokuya bekhathazeka ngokuthi lobu buchwepheshe bunganciphisa ikhono labantu lokucabanga ngokuzimela.
Ucwaningo olwashicilelwa yi-Workday ngo-2025 lubonise ukuthi cishe uhhafu wabaphenduli ukhathazekile ngokuthi i-AI izokwehlisa ikhono labantu lokucabanga ngokuhlolisisa.
Ngesikhathi esifanayo, abantu sebeqalile nokunaka imithelela engaba khona ye-AI empilweni yengqondo. I-OpenAI njengamanje ibhekene namacala amaningi ahlobene nempilo yengqondo yabasebenzisi, kanti amanye awo asola ukuthi i-ChatGPT yake yakhuthaza abasebenzisi ukuthi bazibulale. NgoJuni walo nyaka, izifundazwe eziningi zase-United States zaqala ngokuhlanganyela uphenyo ngomthelela we-ChatGPT entsheni.
Ngaphezu kwalokho, inkinga “yemibono eyiphutha ye-AI” kumamodeli amakhulu olimi, okusho ukukhiqiza okuqukethwe okungalungile noma okungelona iqiniso, iyaqhubeka nokunakwa. Amaphutha anjalo asevele avela emibhalweni yezomthetho efakwe ezinkantolo, okuphoqe abameli namafemu abameli ukuthi baxolise esidlangalaleni; izinkinga ezifanayo ziye zavela nasemibikweni yocwaningo yezinkampani ezinkulu zomhlaba zokubonisana.
UGreen uthe inhloso yakhe enkulu yokusebenzisa i-AI kwakuwukusiza ekutholeni izinto zocwaningo, hhayi ukuqedela ukudala okuqukethwe ngokuqondile. Kodwa wavuma nokuthi izibukeli zinelungelo lokwazi ukuthi yikuphi okuqukethwe okuvela kuye nokuthi yikuphi okuvela ku-AI.
Uthe: “Bengisebenza ngokushesha kakhulu, kangangokuthi angikayiqondi ngokuphelele nenqubo yami yokudala. Ngiyethemba ukuthi ngokuzayo ngizokwazi ukunikeza isiqinisekiso esicacile ngalokhu.”
UGreen uphinde wathi akaphikisani ne-AI, kodwa uyaziqonda izinkathazo zabalandeli abaningi nge-AI, njengendlela amamodeli aqeqeshwa ngayo, umthelela emvelweni, kanye nokuthi izinkampani ezinkulu zobuchwepheshe zisebenzisa lokhu ukuqhubeka nokugxilisa amandla omnotho.
Ekugcineni waxolisa kubasekeli bakhe. Uthe: “Kungenzeka kakhulu ukuthi wena, lowo ofunda lo mbhalo, unemibono ehlukile kweyami mayelana nokuthi i-AI kufanele isetshenziswe yini nokuthi kufanele isetshenziswe kanjani. Ngicabanga ukuthi lokho kwehlukana kwemibono kujwayelekile. Kodwa bengilokhu ngikholelwa ukuthi izibukeli zami zinethuba elikhulu kunami lokuba ziqinisile. Nina niyakwazi ukuzwa ezinye izinto engingakwazi ukuzizwela ngokugcwele. Ngiyaxolisa kakhulu ngoba ngazidedisela kude nenjongo yami yokuqala, futhi ngalimanisa konke esakhe ndawonye.”
UGreen uphinde wathi ubelokhu ebhekene nengcindezi yokudala okuqukethwe ngokuqhubekayo. Esikhathini esizayo uzozama ukuphatha kangcono isimo sakhe somsebenzi, aphinde ahlole ubudlelwano bakhe ne-AI, ngoba lobu budlelwano sebuphenduke “obungenampilo”.
Uthe: “Okubaluleke nakakhulu ukuthi kufanele ngamukele iqiniso lokuthi ukukhuthazeka kwe-dopamine okubangelwa ukuxhumana njalo namamodeli amakhulu olimi nokuzama ukwenza izinto ezengeziwe akulungile kimi, futhi akuwusizi lo mhlaba. Lesi simo singenze ngasebenza ngokunganaki, futhi kancane kancane sangidedisela kude nezinto abantu abazikhathalela ngempela.”