I-OpenAI ivula ucango lwe-ChatGPT ye-erotic engqinisiswe ngobudala

Uhlaziyo lokugqibela: 15/10/2025

  • I-OpenAI iya kuvumela umxholo we-erotic kwi-ChatGPT kubantu abadala abaqinisekisiweyo ukuqala ngoDisemba.
  • Eli nqaku liza kuba ngokuzikhethela, lidityaniswe nokuhlolwa kweminyaka kunye nokhuseleko olutsha.
  • Inguqulelo ye-chatbot enobuntu obusondeleyo kwi-GPT-4 kwaye iqwalaselwe ngumsebenzisi iya kuza.
  • Ibhunga elithi "Wellness and AI" liyadalwa; iinkxalabo malunga nemfihlo kunye nokuqinisekiswa kwesazisi ziyakhula.

Imowudi ye-Erotic kwi-ChatGPT

Kwisibhengezo esibangele ingxoxo, uSam Altman ukuqinisekisile oko I-OpenAI iya kwenza iincoko ezivuselelayo kubantu abadala abaqinisekisiweyo. kwi-ChatGPT eqala ngoDisemba, njengenxalenye yayo umgaqo "wokuphatha abasebenzisi abadala njengabantu abadala"Umlinganiselo uza emva kweenyanga zohlengahlengiso kwimodareyitha kunye nokhuseleko.

Ukongeza, inkampani ibhengeze ukuba iza kuqalisa kungekudala uguqulelo lwe-ChatGPT enobuntu obungakumbi kunye nobuntu obulungisekayo, ukubuyisela iimpawu ezithandwayo malunga nemodeli ye-GPT-4o kwaye, ngokutsho kwe-Altman, yalahleka kunye nemigaqo-nkqubo ethintelweyo egxile kwimpilo yengqondo.

Ukuqinisekiswa kweminyaka yobudala kunye nokufikelela kwimo evuselela inkanuko

Incoko ye-GPT ye-Erotic kunye noQinisekiso lobudala

I-OpenAI ibophelela olu tshintsho kwi-a inkqubo yokulawula iminyaka eya kuchonga kwaye iqinisekise iminyaka yoninzi phambi kokwenza umxholo "omdala".Inketho ayizukwenziwa ngokungagqibekanga kwaye iya kuvela kuphela xa umsebenzisi eyicela kwaye ihlangabezana neemfuno zokuqinisekisa.

Inkampani yayisele ikhankanye ukuba iza kuvumela abaphuhlisi yenza usetyenziso olunomxholo omdala (18+) ngaphakathi kwe-ChatGPT ecosystem, ngaphandle kokuba isiqinisekiso kunye iindlela zokulawula zaphunyezwa ngokupheleleyo. Eli nyathelo ngoku liyenzeka ngokwandiswa kwemida kubasebenzisi abadala.

Umxholo okhethekileyo- Cofa Apha  ILenovo ibonisa iiglasi zayo ze-AI eziBonakalayo iiglasi ze-AI ze-V1

Okungaziwa okufanelekileyo kushiyekile: I-OpenAI ayikhange ichaze eyona ndlela yokuqinisekisa. okanye ukuba uxwebhu lwesazisi, i-selfie okanye ungqinisiso oluxutyiweyo luya kufuneka. Phakathi koluntu kuye kwavela iinkxalabo zabucala, kunye nabasebenzisi abangafuni ukulayisha idatha ebuthathaka nokuba isithembiso kukuphucula ubuqu kunye namava.

U-Altman akawuchazanga umda opheleleyo wendlela evuselela umdla; okwangoku, umyalezo osemthethweni ugxile iincoko ezinemixholo yabantu abadala phantsi kolawulo olungqongqo, ngaphandle kwentengiso yee-avatar ezicacileyo okanye amava abonwayo. Isikhokelo sesengxoxo kunye nesicatshulwa kubantu abadala, kunye nezikhuselo ezisebenzayo.

Izikhuselo kunye nempilo yengqondo: izithintelo ezimbalwa, ulawulo oluninzi

U-Altman ngokwakhe uvumile ukuba, kwezi nyanga zimbalwa zidlulileyo, I-ChatGPT "yayingqongqo kakhulu" xa ujongene neentlekele zempilo yengqondo, imeko enxulumene ne ukususa izilumkiso kwi-ChatGPT, Intoni Kunciphise ukuba luncedo kunye nokonwabela inxalenye yesiseko somsebenzisi. ngaphandle olo hlobo lomngcipheko.

Ngokutsho kwe-CEO, inkampani isasazekile izixhobo zokubona ngcono uxinzelelo lweemvakalelo kunye nezinye iimpawu zomngcipheko, ezinokuvumela "ukuphumla okukhuselekileyo kwezithintelo kwiimeko ezininzi." Olu tshintsho lukhatshwa a iingcebiso "kwintlalontle kunye ne-AI" kunye nabaphandi bagxile kwimpembelelo yeteknoloji kwimpilo yengqondo, nangona kuye kwaphawulwa ukuba ayibandakanyi iingcali zokuthintela ukuzibulala.

Umxholo okhethekileyo- Cofa Apha  Konke malunga noCofa ukwenza: Ukwenziwa kwezinto ezintsha Windows 11 kwiscreen sakho

Ingxoxo ayincinci: I-OpenAI ijongene nayo Ukugxekwa kunye nesimangalo esinxulumene netyala lolutsha, umxholo oye wakhuthaza ukuqiniswa kwezikhuselo, iiprothokholi zokuphendula, kunye nemida yokuziphatha kwiimeko ezibuthathaka. Inkampani igxininisa ukuba ukhuseleko luya kuhlala luphambili.

Umntu "omntu" ngakumbi kunye nobuntu obulungelelanisiweyo, isitayela se-GPT-4o

I-Erotic GPT Incoko

"Kwiiveki ezimbalwa," i-OpenAI iceba ukusungula uhlaziyo oluya kukuvumela ukuba ulungise ubuntu bomncedisi: iimpendulo zendalo ngaphezulu, ukusetyenziswa kwe-emojis okanye ithoni esondele ngakumbi, rhoqo "ukuba umsebenzisi uyayifuna." Umbono kukubuyisela iimpawu ukuba ezininzi ezinxulumene ne-GPT-4ongelixa ugcina isakhelo sokhuseleko esomeleleyo.

Olu hlengahlengiso lujolise ekusombululeni into yokuba, emehlweni abanye, i-chatbot Wabanda, ezibamba ngokugqithiseleyo yaye engazenzekeli.I-OpenAI ithembisa ukulinganisela: ukufudumala okungaphezulu kunye nokuguquguquka ekusebenzisaneni nabanye, kodwa kunye nemida ecacileyo kunye nokukwazi ukuthintela xa imeko ifuna.

Ukhuphiswano kunye nentengiso: I-xAI sele idlala ngothando namava abantu abadala

I-OpenAI ayisebenzi kwi-vacuum. xAI, inkampani kaElon Musk, uye wazama nge-anime-style chatbots kwi-Grok eyenzelwe "ukudlala ngothando," umgca obonisa imfuno yamava encoko asondeleyo kwimarike

Umxholo okhethekileyo- Cofa Apha  Iividiyo zikaGoogle: Ukuhlelwa kwevidiyo ngqo kwiDrive

Umahluko, okwangoku, yindlela: I-OpenAI igxininisa Ukuqinisekiswa kweminyaka yobudala, ulawulo, kunye nokungabikho kwee-avatar ezicacileyo, ukubeka phambili ukuthotyelwa kunye nokhuseleko. Ukuvula imowudi evuselela inkanuko kwi-ChatGPT kunokukhawulezisa umdyarho iinkonzo zabantu abadala ezinemiqobo yokhuseleko engaphezulu kunye nemigangatho ecacileyo.

Yintoni eseleyo ukuba ichazwe

Inguqulelo ye-Erotic ye-ChatGPT

U-Altman ubeka ukwandiswa EyoMnga, edityaniswe “neyona nto igqibeleleyo” yokuphunyezwa kolawulo lobudala, kodwa kungekho mhla wokugqibela okanye uludwe olucacileyo lwamazwe apho luza kuqala lusebenze. Kwakhona kusaza kubonakala ukuba kuya kubakho ezinye iindlela zokuqinisekisa kwabo bangafuni ukufaka amaxwebhu anovakalelo.

Ngokuhambelanayo, kuya kufuneka sibone ukuba olu tshintsho lungena njani kwi Imigaqo yokusetyenziswa, ukumodareyitha, kunye nemigaqo-nkqubo yophuhlisiUmceli mngeni uya kuba kukugcina isithembiso "sokuphatha abantu abadala njengabantu abadala" ngaphandle kokuphumla i-guardrails ekhusela abancinci kunye nabasebenzisi abasesichengeni.

Kusalindelwe iinkcukacha zokugqibela, intshukumo ye-OpenAI ijolise kwiChatGPT ebhetyebhetye kwaye enokwenzeka ngokwezifiso: ukukwazi ukwamkela ithoni yomntu ngakumbi, ulawulo lokujonga iimeko ezinobuzaza, kunye nemo evuselela inkanuko engqinisiswe ngobudala. Konke oku kuza noxinzelelo oluvela kwimarike esele iphonononga amava asondeleyo nge-AI kunye neglasi yokukhulisa. ubumfihlo, ukhuseleko kunye noxanduva.

I-Reddit imangalela i-Anthropic
Inqaku elidibeneyo:
I-Reddit imangalela i-Anthropic ngokusetyenziswa okungagunyaziswanga kwedatha yayo kwi-AI