Yini i-AI ekhiqizayo engathenjelwa kuyo ngaphandle kokungenelela komuntu?

Isifinyezo esiphezulu

I-Generative Artificial Intelligence (AI) – ubuchwepheshe obuvumela imishini ukudala umbhalo, izithombe, ikhodi, nokuningi – bukhule kakhulu eminyakeni yamuva nje. Leli phepha elimhlophe linikeza umbono ofinyelelekayo walokho i-AI ekhiqizayo engakwenza ngokwethembeka namuhla ngaphandle kokungenela komuntu, nokuthi kulindeleke ukuthi yenzeni eminyakeni eyishumi ezayo. Sihlola ukusetshenziswa kwayo ekubhaleni, ubuciko, ukubhala amakhodi, insizakalo yamakhasimende, ukunakekelwa kwempilo, imfundo, ezokuthutha, kanye nezezimali, siqokomisa lapho i-AI isebenza ngokuzimela khona nokuthi lapho ukubhekwa kwabantu kuhlala kubalulekile khona. Izibonelo zangempela zifakiwe ukuze zibonise impumelelo kanye nemikhawulo. Okutholakele okuyinhloko kufaka phakathi:

• Ukwamukelwa Okubanzi: Ngo-2024, ama-65% ezinkampani ezihloliwe abika njalo ukuthi zisebenzisa i-AI ekhiqizayo - cishe isabelo esiphindwe kabili kunesonyaka odlule ( Isimo se-AI ekuqaleni kuka-2024 | McKinsey ). Izinhlelo zokusebenza zihlanganisa ukudalwa kokuqukethwe kokumaketha, ama-chatbot okusekela amakhasimende, ukukhiqizwa kwekhodi, nokuningi.

• Amakhono Amanje Azimele: I-AI yokukhiqiza yanamuhla iphatha imisebenzi ehlelekile nephindaphindwayo ngaphandle kokuqapha okuncane. Izibonelo zifaka phakathi ukukhiqiza ngokuzenzakalelayo imibiko yezindaba ehlelekile (isb. izifinyezo zenzuzo yenkampani) ( Philana Patterson – Iphrofayela Yomphakathi ye-ONA ), ukukhiqiza izincazelo zomkhiqizo kanye nokugqama kokubuyekezwa kumasayithi e-commerce, kanye nekhodi yokuqedela ngokuzenzakalelayo. Kulezi zizinda, i-AI ivame ukukhulisa abasebenzi abangabantu ngokuthatha izintambo zokukhiqiza okuqukethwe okuvamile.

• I-Human-in-the-Loop Yemisebenzi Eyinkimbinkimbi: Emisebenzini eyinkimbinkimbi noma evulekile - njengokubhala kokudala, ukuhlaziya okuningiliziwe, noma iseluleko sezokwelapha - ukugadwa kwabantu kusadingeka ukuqinisekisa ukunemba kwamaqiniso, ukwahlulela kokuziphatha, kanye nekhwalithi. Ukufakwa okuningi kwe-AI namuhla kusebenzisa imodeli "yomuntu-in-the-loop" lapho i-AI idweba khona okuqukethwe bese abantu bekubuyekeza.

• Ukuthuthukiswa Kwesikhathi Esiseduze: Eminyakeni emi-5-10 ezayo, i-AI yokukhiqiza kulindeleke ukuthi ibe nokwethenjelwa kakhulu futhi izimele . Intuthuko ekunembileni kwemodeli kanye nezindlela zokuqapha ingase ivumele i-AI ukuthi iphathe ingxenye enkulu yemisebenzi yokudala nokwenza izinqumo ngokufakwa okuncane kwabantu. Isibonelo, ngo-2030 ochwepheshe babikezela ukuthi i-AI izophatha iningi lokuxhumana kwenkonzo yamakhasimende kanye nezinqumo ngesikhathi sangempela ( Ukucabanga Kabusha Ngokushintsha Ushintsho Ku-CX, Abamakethi Kumelwe Bakwenze Lezi Zinto Ezi-2 ), futhi ifilimu enkulu ingakhiqizwa ngokuqukethwe okukhiqizwe yi-AI okungu-90% ( Amacala Okusetshenziswa Kwe-AI Okukhiqizayo Kwamabhizinisi Namabhizinisi ).

• Ngo-2035: Eminyakeni eyishumi, silindele ukuthi ama-AI azimele abe yinto evamile emikhakheni eminingi. Abafundisi be-AI banganikeza imfundo yomuntu siqu ngezinga elithile, abasizi be-AI bangase babhale ngokuthembekile izinkontileka zomthetho noma imibiko yezokwelapha ukuze kusayinwe ochwepheshe, kanti izinhlelo zokuzishayela (ezisizwa ukulingisa okukhiqizayo) zingase ziqhube imisebenzi yezokuthutha kusukela ekuqaleni kuze kube sekupheleni. Kodwa-ke, izindawo ezithile ezibucayi (isib. ukuxilongwa kwezokwelapha okubaluleke kakhulu, izinqumo zomthetho zokugcina) cishe zizodinga ukwahlulela komuntu ngokuphepha nokuziphendulela.

• Ukukhathazeka Ngokuziphatha Nokuthembeka: Njengoba ukuzimela kwe-AI kukhula, nokukhathazeka kuyakhula. Izinkinga namuhla zifaka phakathi ukubona izinto ezingekho (ukuqamba amaqiniso nge-AI), ukukhetha kokuqukethwe okukhiqizwe, ukuntuleka kokubonakala, kanye nokusetshenziswa kabi kolwazi olungamanga. Ukuqinisekisa ukuthi i-AI ingathenjwa uma isebenza ngaphandle kokubhekwa kubaluleke kakhulu. Intuthuko iyenziwa - isibonelo, izinhlangano zitshala imali eningi ekunciphiseni ubungozi (ukubhekana nokunemba, ukuphepha kwe-cyber, izinkinga ze-IP) ( Isimo se-AI: Ucwaningo Lomhlaba Wonke | McKinsey ) - kodwa kudingeka ukubusa okuqinile kanye nezinhlaka zokuziphatha.

• Isakhiwo saleli Phepha: Siqala ngesingeniso se-AI yokukhiqiza kanye nomqondo wokusetshenziswa okuzenzakalelayo vs. okuqondisiwe. Bese kuthi, kuzo zonke izigaba ezinkulu (ukubhala, ubuciko, ukubhala amakhodi, njll.), sixoxe ngalokho i-AI engakwenza ngokwethembeka namuhla uma kuqhathaniswa nalokho okuseduze. Siphetha ngezinselele ezihlanganisa konke, izilinganiso zesikhathi esizayo, kanye nezincomo zokusebenzisa i-AI yokukhiqiza ngokuzibophezela.

Sekukonke, i-AI ekhiqizayo isivele ibonakalise ukuthi iyakwazi ukusingatha imisebenzi eminingi emangalisayo ngaphandle kwesiqondiso somuntu esiqhubekayo. Ngokuqonda imikhawulo yayo yamanje kanye namandla ayo esikhathi esizayo, izinhlangano kanye nomphakathi bangalungiselela kangcono inkathi lapho i-AI ingeyona nje ithuluzi, kodwa futhi ingumhlanganyeli ozimele emsebenzini nasebucikweni.

Isingeniso

Ubuhlakani Bokwenziwa bukwazile isikhathi eside ukuhlaziya idatha, kodwa muva nje izinhlelo ze-AI zifunde ukudala - ukubhala iphrozi, ukubhala izithombe, isofthiwe yokuhlela, nokuningi. Lawa e-AI akhiqizayo (njenge-GPT-4 yombhalo noma i-DALL·E yezithombe) aqeqeshwe kumasethi edatha amakhulu ukukhiqiza okuqukethwe okusha ukuphendula izimpendulo. Lokhu kuphumelela kuveze igagasi lokusungula izinto ezintsha kuzo zonke izimboni. Kodwa-ke, kuphakama umbuzo obalulekile: Yini esingayethemba ngempela ukuthi i-AI izoyenza yodwa, ngaphandle kokuthi umuntu ahlole kabili umphumela wayo?

Ukuze uphendule lokhu, kubalulekile ukuhlukanisa phakathi AI okuqondisiwe kanye okuzenzakalelayo kwe-AI:

• I-AI eqondiswa abantu ibhekisela ezimweni lapho imiphumela ye-AI ibukezwa noma ihlelwa ngabantu ngaphambi kokuba iqedwe. Isibonelo, intatheli ingasebenzisa umsizi wokubhala we-AI ukubhala isihloko, kodwa umhleli uyasihlela futhi asivume.

• I-Autonomous AI (i-AI ngaphandle kokungenela komuntu) ibhekisela ezinhlelweni ze-AI ezenza imisebenzi noma ezikhiqiza okuqukethwe okusetshenziswa ngqo ngaphandle kokuhlela okuncane noma okungekho nhlobo komuntu. Isibonelo i-chatbot ezenzakalelayo exazulula umbuzo wamakhasimende ngaphandle kwe-ejenti yomuntu, noma isikhungo sezindaba esishicilela ngokuzenzakalelayo isifinyezo samaphuzu ezemidlalo esikhiqizwe yi-AI.

I-AI Ekhiqizayo isivele isetshenziswa kuzo zombili izindlela. Ngo-2023-2025, ukwamukelwa kwayo kukhuphuke kakhulu , kanti izinhlangano ziyazama ngentshiseko. Ucwaningo olulodwa lomhlaba wonke ngo-2024 luthole ukuthi ama-65% ezinkampani asebenzisa njalo i-AI ekhiqizayo, kusukela cishe engxenyeni eyodwa kwezintathu ngonyaka odlule ( Isimo se-AI ekuqaleni kuka-2024 | McKinsey ). Abantu ngabanye nabo basebenzise amathuluzi afana ne-ChatGPT - kulinganiselwa ukuthi ama-79% ochwepheshe babe nokuchayeka okungenani okuthile ku-AI ekhiqizayo maphakathi no-2023 ( Isimo se-AI ngo-2023: Unyaka wokuqhamuka kwe-AI ekhiqizayo | McKinsey ). Lokhu kutholwa okusheshayo kuqhutshwa yisithembiso sokusebenza kahle kanye nokuzuza kokudala. Kodwa-ke kusalokhu "kusezinsukwini zokuqala," futhi izinkampani eziningi zisakha izinqubomgomo zendlela yokusebenzisa i-AI ngokuzibophezela ( Isimo se-AI ngo-2023: Unyaka wokuqhamuka kwe-AI ekhiqizayo | McKinsey ).

Kungani ukuzimela kubalulekile: Ukuvumela i-AI isebenze ngaphandle kokuqondisa komuntu kungavula izinzuzo ezinkulu zokusebenza kahle - ukwenza imisebenzi eyisicefe ngokuphelele - kodwa futhi kuphakamisa ukuthembeka. I-ejenti ye-AI ezimele kumele ilungise izinto (noma yazi imingcele yayo) ngoba kungenzeka kungabikho muntu ngesikhathi sangempela obamba amaphutha. Eminye imisebenzi ivumela lokhu kakhulu kuneminye. Ngokuvamile, i-AI isebenza kahle kakhulu ngokuzimela uma:

• Umsebenzi unesakhiwo noma iphethini ecacile (isb. ukukhiqiza imibiko ejwayelekile kusuka kudatha).

• Amaphutha ayingozi encane noma abekezelelwa kalula (isb. ukukhiqizwa kwesithombe okungalahlwa uma kungagculisi, uma kuqhathaniswa nokuxilongwa kwezokwelapha).

• Kukhona idatha yokuqeqesha ehlanganisa izimo, ngakho-ke umphumela we-AI usekelwe ezibonelweni zangempela (ukunciphisa ukuqagela).

Ngokuphambene nalokho, imisebenzi evulekile , enezinselelo ezinkulu , noma edinga ukwahlulela okucacile ayifaneleki kangako ekungaqondisweni nhlobo namuhla.

Ezigabeni ezilandelayo, sihlola uhla lwezinkambu ukuze sibone ukuthi i-AI ekhiqizayo yenzani manje nokuthi yini elandelayo. Sizobheka izibonelo eziqondile - kusukela ezihlokweni zezindaba ezibhalwe yi-AI kanye nobuciko obukhiqizwe yi-AI, kuya kubasizi bokubhala amakhodi kanye nama-ejenti enkonzo yamakhasimende abonakalayo - siqokomisa ukuthi yimiphi imisebenzi engenziwa kusukela ekuqaleni kuya ekugcineni yi-AI futhi okusadinga umuntu ohlala ekhona. Kwisizinda ngasinye, sihlukanisa ngokucacile amakhono amanje (cishe ngo-2025) ekubikezelweni okungokoqobo kwalokho okungaba okuthembekile ngo-2035.

Ngokuhlela imephu yamanje nekusasa le-AI ezimele kuzo zonke izizinda, sihlose ukunikeza abafundi ukuqonda okulinganiselayo: singayithathi ngokweqile i-AI njengengenaphutha ngokumangalisayo, noma siyinciphise amakhono ayo angempela nakhulayo. Ngalesi sisekelo, sixoxa ngezinselele ezibanzi zokuthembela ku-AI ngaphandle kokugadwa, okuhlanganisa nokucatshangelwa kokuziphatha kanye nokuphathwa kwezingozi, ngaphambi kokuphetha ngezinto ezibalulekile.

I-AI Ekhiqizayo Ekubhaleni Nasekudalweni Kokuqukethwe

Esinye sezizinda zokuqala lapho i-AI ekhiqizayo yathola khona udumo kwakuwukukhiqizwa kombhalo. Amamodeli amakhulu olimi angakhiqiza konke kusukela ezihlokweni zezindaba kanye nekhophi yokumaketha kuya kokuthunyelwe ezinkundleni zokuxhumana kanye nezifinyezo zamadokhumenti. Kodwa kungakanani lokhu kubhala okungenziwa ngaphandle komhleli ongumuntu?

Amakhono Amanje (2025): I-AI njengoMbhali Ozenzakalelayo Wokuqukethwe Okuvamile

Namuhla, i-AI yokukhiqiza iphatha imisebenzi ehlukahlukene yokubhala evamile ngaphandle kokungenelela komuntu noma ngaphandle kokungenelela komuntu. Isibonelo esiyinhloko sikubuntatheli: i-Associated Press iminyaka eminingi isebenzise ukuzenzakalela ukukhiqiza izinkulungwane zemibiko yenzuzo yenkampani njalo ngekota ngqo kusuka ku-data feeds ( Philana Patterson - ONA Community Profile ). Lezi zingcezu zezindaba ezimfushane zilandela ithempulethi (isb., “Inkampani X ibike inzuzo engu-Y, ikhuphuke ngo-Z%...”) kanti i-AI (esebenzisa isofthiwe yokukhiqiza ulimi lwemvelo) ingagcwalisa izinombolo kanye namazwi ngokushesha kunanoma yimuphi umuntu. Uhlelo lwe-AP lushicilela le mibiko ngokuzenzakalelayo, lwandisa ukumbozwa kwayo kakhulu (ngaphezu kwezindaba ezingu-3,000 ngekota) ngaphandle kokudinga ababhali abangabantu ( Izindaba zenzuzo ezizenzakalelayo ziyanda | I-Associated Press ).

Ubuntatheli bezemidlalo bukhuliswe ngendlela efanayo: Izinhlelo ze-AI zingathatha izibalo zemidlalo yezemidlalo futhi zikhiqize izindaba ezifingqiwe. Ngenxa yokuthi lezi zizinda ziqhutshwa idatha futhi ziyindlela yokwenza izinto, amaphutha awavamile uma nje idatha ilungile. Kulezi zimo, sibona ukuzimela kwangempela - i-AI iyabhala bese okuqukethwe kushicilelwa ngokushesha.

Amabhizinisi asebenzisa ne-AI ekhiqizayo ukubhala izincazelo zomkhiqizo, izincwadi zezindaba ze-imeyili, kanye nokunye okuqukethwe kokumaketha. Isibonelo, inkampani enkulu yezentengiselwano ze-e-commerce i-Amazon manje isebenzisa i-AI ukufingqa izibuyekezo zamakhasimende zemikhiqizo. I-AI iskena umbhalo wezibuyekezo eziningi ngazinye futhi ikhiqize isigaba esifingqiwe salokho abantu abakuthandayo noma abangakuthandi ngento, esibe sesiboniswa ekhasini lomkhiqizo ngaphandle kokuhlelwa ngesandla ( i-Amazon ithuthukisa ulwazi lokubuyekezwa kwamakhasimende nge-AI ). Ngezansi umfanekiso walesi sici esisetshenziswa kuhlelo lokusebenza lweselula lwe-Amazon, lapho isigaba esithi “Amakhasimende athi” sikhiqizwa ngokuphelele yi-AI kusuka kudatha yokubuyekezwa:

( I-Amazon ithuthukisa ulwazi lokubuyekezwa kwamakhasimende nge-AI ) isifinyezo sokubuyekezwa okukhiqizwe yi-AI ekhasini lomkhiqizo we-e-commerce. Uhlelo lwe-Amazon lufingqa amaphuzu avamile avela kuzibuyekezo zabasebenzisi (isb., ukusebenziseka kalula, ukusebenza) esigabeni esifushane, esiboniswa kubathengi njenge-“AI ekhiqizwe kusukela embhalweni wezibuyekezo zamakhasimende.”

Amacala okusetshenziswa anjalo abonisa ukuthi uma okuqukethwe kulandela iphethini ebikezelwayo noma kuhlanganiswa kusuka kudatha ekhona, i-AI ivame ukuyisingatha yodwa . Ezinye izibonelo zamanje zifaka:

• Izibuyekezo Zesimo Sezulu Nethrafikhi: Iziteshi zemidiya ezisebenzisa i-AI ukuqoqa imibiko yesimo sezulu yansuku zonke noma ama-bulletin ethrafikhi ngokusekelwe kudatha yenzwa.

• Imibiko Yezezimali: Izinkampani ezikhiqiza izifinyezo zezimali eziqondile (imiphumela yekota, izifinyezo zemakethe yamasheya) ngokuzenzakalelayo. Kusukela ngo-2014, iBloomberg nezinye izitolo zezindaba zisebenzise i-AI ukusiza ekubhaleni izincazelo zezindaba ngemali engenayo yenkampani - inqubo esebenza ngokuzenzakalelayo kakhulu uma idatha isifakiwe (' izintatheli zerobhothi' ze-AP zibhala izindaba zazo manje | I-Verge ) ( Intatheli yaseWyoming ibanjwe isebenzisa i-AI ukudala izingcaphuno ezingamanga, izindaba ).

• Ukuhumusha Nokubhala: Izinsizakalo zokubhala manje zisebenzisa i-AI ukukhiqiza imibhalo noma imibhalo engezansi yomhlangano ngaphandle kwabantu ababhala ngomshini. Nakuba ingeyona imisebenzi yokudala ngomqondo wokudala, le misebenzi yolimi isebenza ngokuzenzakalelayo ngokunemba okuphezulu ukuze kube nomsindo ocacile.

• Ukukhiqiza Okusalungiswa: Ochwepheshe abaningi basebenzisa amathuluzi afana ne-ChatGPT ukubhala ama-imeyili noma izinguqulo zokuqala zamadokhumenti, ngezinye izikhathi bewathumela ngaphandle kokuhlelwa okuningi noma okungekho uma okuqukethwe kungeyona ingozi enkulu.

Kodwa-ke, ukuze kube nephrozi eyinkimbinkimbi kakhulu, ukuqondiswa kwabantu kusalokhu kuyinto evamile ngo-2025 . Izinhlangano zezindaba azivamile ukushicilela izihloko eziphenyayo noma ezihlaziyayo ngqo kusuka ku-AI - abahleli bazohlola amaqiniso futhi balungise iziqephu ezibhalwe yi-AI. I-AI ingalingisa isitayela nesakhiwo kahle kodwa ingaletha amaphutha amaqiniso (ngokuvamile abizwa ngokuthi “ama-hallucinations”) noma imisho exakile umuntu okudingeka ayibambe. Isibonelo, iphephandaba laseJalimane i-Express lethula “umlingani wedijithali” we-AI ogama lakhe linguKlara ukusiza ekubhaleni izingcezu zezindaba zokuqala. I-Klara ingabhala kahle imibiko yezemidlalo futhi ibhale nezihloko eziheha abafundi, okunegalelo ku-11% wezihloko ze-Express - kodwa abahleli babantu basabuyekeza yonke into ukuze bathole ukunemba nobuqotho bezintatheli, ikakhulukazi ezindabeni eziyinkimbinkimbi ( Izindlela Eziyi-12 Izintatheli Ezisebenzisa Amathuluzi E-AI Egumbini Lezindaba - Twipe ). Lobu budlelwano phakathi kwabantu ne-AI buvamile namuhla: I-AI iphatha ukukhushulwa okunzima kokukhiqiza umbhalo, futhi abantu bayalungisa futhi balungise njengoba kudingeka.

Umbono Wonyaka Ka-2030-2035: Ukubhala Okuzimele Okuthembekile

Eminyakeni eyishumi ezayo, silindele ukuthi i-AI ekhiqizayo ithembeke kakhulu ekukhiqizeni umbhalo osezingeni eliphezulu, onembile, ozokwandisa ububanzi bemisebenzi yokubhala engayisingatha ngokuzimela. Izitayela eziningana zisekela lokhu:

• Ukunemba Okuthuthukisiwe: Ucwaningo oluqhubekayo lunciphisa ngokushesha ukuthambekela kwe-AI kokukhiqiza ulwazi olungamanga noma olungafanele. Ngo-2030, amamodeli olimi athuthukile anokuqeqeshwa okungcono (kufaka phakathi amasu okuqinisekisa amaqiniso ngokumelene nezizindalwazi ngesikhathi sangempela) angafeza ukuhlolwa kwamaqiniso okucishe kube sezingeni lomuntu ngaphakathi. Lokhu kusho ukuthi i-AI ingase ibhale isihloko sezindaba esigcwele ngezingcaphuno nezibalo ezifanele ezithathwe ezintweni zomthombo ngokuzenzakalelayo, okudinga ukuhlelwa okuncane.

• Ama-AI Akhethekile Esizinda: Sizobona amamodeli akhethekile okukhiqiza ahlelwe kahle emikhakheni ethile (ezomthetho, ezokwelapha, nokubhala kobuchwepheshe). Imodeli ye-AI yezomthetho ka-2030 ingase ibhale ngokuthembekile izinkontileka ezijwayelekile noma ifinyeze umthetho wamacala - imisebenzi enesakhiwo esihlelekile kodwa okwamanje idinga isikhathi sommeli. Uma i-AI iqeqeshwe kumadokhumenti asemthethweni aqinisekisiwe, ama-draft ayo angase athembeke ngokwanele ukuthi ummeli unikeza umbono wokugcina ngokushesha.

• Isitayela Semvelo Nokuhambisana: Amamodeli ayathuthuka ekugcineni umongo ngaphezu kwamadokhumenti amade, okuholela kokuqukethwe okunemininingwane eminingi nokuqondile. Ngo-2035, kungenzeka ukuthi i-AI ingabhala uhlaka lokuqala oluhle lwencwadi engeyona inganekwane noma incwadi yezobuchwepheshe ngokwayo, abantu ikakhulukazi besendimeni yokweluleka (ukubeka imigomo noma ukunikeza ulwazi olukhethekile).

Lokhu kungase kubukeke kanjani empeleni? Ubuntatheli obujwayelekile bungase bube cishe ngokuzenzakalelayo ngokuphelele kwezinye izingoma. Singase sibone i-ejensi yezindaba ngo-2030 inohlelo lwe-AI olubhala inguqulo yokuqala yombiko ngamunye wemali engenayo, indaba yezemidlalo, noma isibuyekezo semiphumela yokhetho, kanye nomhleli othatha isampula embalwa kuphela ukuze kuqinisekiswe ikhwalithi. Ngempela, ochwepheshe babikezela ukuthi isabelo esikhula njalo sokuqukethwe okuku-inthanethi sizokhiqizwa ngomshini - esinye isibikezelo esinesibindi sabahlaziyi bemboni siphakamise ukuthi kufike ku-90% wokuqukethwe okuku-inthanethi kungakhiqizwa yi-AI ngo-2026 ( Ngo-2026, Okuqukethwe Okuku-inthanethi Okukhiqizwa Ngabantu Abangebona Abantu Kuzodlula Kakhulu Okuqukethwe Okukhiqizwa Abantu - OODAloop ), yize lelo nani liphikiswana ngalo. Ngisho nomphumela olondolozayo ungasho ukuthi maphakathi nawo-2030, iningi lezihloko zewebhu ezijwayelekile, ikhophi yomkhiqizo, kanye nokuphakelwa kwezindaba okwenziwe ngezifiso kubhalwe yi-AI.

Ekukhangiseni nasekuxhumaneni kwezinkampani , i-AI ekhiqizayo cishe izophathiswa ukuqhuba imikhankaso yonke ngokuzimela. Ingakhiqiza futhi ithumele ama-imeyili okumaketha enziwe ngokwezifiso, okuthunyelwe ezinkundleni zokuxhumana, kanye nokwehluka kwamakhophi ezikhangiso, ishintsha njalo imiyalezo ngokusekelwe ekuphenduleni kwamakhasimende - konke ngaphandle kombhali wekhophi ongumuntu onolwazi. Abahlaziyi be-Gartner babikezela ukuthi ngo-2025, okungenani ama-30% emiyalezo yokumaketha yamabhizinisi amakhulu izokhiqizwa ngokwenziwa yi-AI ( I-Generative AI Use Cases for Industries and Enterprises ), futhi leli phesenti lizokhuphuka kuphela ngo-2030.

Kodwa-ke, kubalulekile ukuqaphela ukuthi ubuhlakani bomuntu kanye nokwahlulela kuzoqhubeka nokudlala indima, ikakhulukazi kokuqukethwe okubalulekile . Ngo-2035, i-AI ingase iphathe isitatimende sabezindaba noma okuthunyelwe kwebhulogi ngokwayo, kodwa ngobuntatheli obuphenyayo obuhilela ukuziphendulela noma izihloko ezibucayi, iziteshi zabezindaba zingase zisagcizelela ekuqondisweni komuntu. Ikusasa cishe lizoletha indlela ehlukene: i-AI ikhiqiza ngokuzimela okuqukethwe okuningi kwansuku zonke, kuyilapho abantu begxila ekuhleleni nasekukhiqizeni izingcezu ezinamasu noma ezibucayi. Empeleni, umugqa walokho okubalwa “njengomkhuba” uzokhula njengoba ubuchwepheshe be-AI bukhula.

Ngaphezu kwalokho, izinhlobo ezintsha zokuqukethwe njengezindaba ezisebenzisanayo ezikhiqizwe yi-AI noma imibiko eyenzelwe wena zingavela. Isibonelo, umbiko wonyaka wenkampani ungadalwa ngezindlela eziningi yi-AI - isifinyezo sabaphathi, inguqulo yokulandisa yabasebenzi, inguqulo ecebile ngedatha yabahlaziyi - ngayinye idalwe ngokuzenzakalelayo kusuka kudatha efanayo eyisisekelo. Kwezemfundo, izincwadi zingabhalwa ngokuguquguqukayo yi-AI ukuze zivumelane namazinga ahlukene okufunda. Lezi zinhlelo zokusebenza zingaba ezizimele kakhulu kodwa zisekelwe ulwazi oluqinisekisiwe.

Indlela yokubhala iphakamisa ukuthi maphakathi nawo-2030, i-AI izoba ngumlobi onolwazi oluningi . Isihluthulelo sokusebenza ngokuzimela ngempela kuzoba ukwakha ukwethembana emiphumeleni yayo. Uma i-AI ingabonisa njalo ukunemba kwamaqiniso, ikhwalithi yesitayela, kanye nokuhambisana nezindinganiso zokuziphatha, isidingo sokubuyekezwa komuntu ngokulandelana sizoncipha. Izingxenye zaleli phepha elimhlophe ngokwalo, ngo-2035, zingase zibhalwe ngumcwaningi we-AI ngaphandle kokudinga umhleli - i-prospect esinethemba ngayo ngokucophelela, uma nje kukhona izindlela zokuvikela ezifanele.

I-AI Ekhiqizayo kubuciko bokubuka kanye nokuklama

Ikhono le-AI elikhiqizayo lokudala izithombe nemisebenzi yobuciko lidonse umcabango womphakathi, kusukela emidwebweni eyenziwe yi-AI ewina imincintiswano yobuciko kuya kumavidiyo ajulile angahlukaniseki namavidiyo angempela. Ezindaweni ezibonakalayo, amamodeli e-AI afana namanethiwekhi aphikisanayo akhiqizayo (ama-GAN) kanye namamodeli okusabalalisa (isb. i-Stable Diffusion, i-Midjourney) angakhiqiza izithombe zokuqala ngokusekelwe ezicelweni zombhalo. Ngakho-ke, ingabe i-AI manje ingasebenza njengomculi noma umklami ozimele?

Amakhono Amanje (2025): I-AI njengoMsizi Wokudala

Kusukela ngo-2025, amamodeli akhiqizayo anekhono lokudala izithombe lapho zifunwa ngokwethembeka okumangalisayo. Abasebenzisi bangacela isithombe se-AI ukuthi sidwebe "idolobha lasendulo ekushoneni kwelanga ngesitayela sikaVan Gogh" futhi bathole isithombe sobuciko esikholisayo ngemizuzwana. Lokhu kuholele ekusetshenzisweni kabanzi kwe-AI ekwakhiweni kwezithombe, ukumaketha, kanye nokuzijabulisa kobuciko bomqondo, ama-prototypes, ngisho nezithombe zokugcina kwezinye izimo. Okuphawulekayo:

• Umklamo Wezithombe Nezithombe Zesitoko: Izinkampani zikhiqiza ihluzo zewebhusayithi, imidwebo, noma izithombe zesitoko nge-AI, okunciphisa isidingo sokusebenzisa yonke into evela kumculi. Amaqembu amaningi okumaketha asebenzisa amathuluzi e-AI ukukhiqiza izinhlobo ezahlukene zezikhangiso noma izithombe zomkhiqizo ukuze ahlole ukuthi yini ekhanga abathengi.

• Ubuciko Nokudweba: Abaculi ngabanye basebenzisana ne-AI ukuze bacabange ngemibono noma bagcwalise imininingwane. Isibonelo, umdwebi angasebenzisa i-AI ukukhiqiza indawo yangemuva, abese eyihlanganisa nabalingiswa bakhe abadwetshwe ngabantu. Abanye abadali bezincwadi zamahlaya baye bazama ngamaphaneli noma imibala ekhiqizwe yi-AI.

• Imidiya Nokuzijabulisa: Ubuciko obukhiqizwe yi-AI buvele ezimbozweni zomagazini nasemaphepheni ezincwadi. Isibonelo esidumile kwakuyisembozo -Cosmopolitan esasibonisa usomkhathi - okubikwa ukuthi kwakuyisithombe sesembozo sokuqala sikamagazini esadalwa yi-AI (i-OpenAI's DALL·E) njengoba kuqondiswe umqondisi wezobuciko. Nakuba lokhu kwakuhilela ukukhuthazwa nokukhetha kwabantu, umsebenzi wobuciko wangempela wawukhiqizwa ngomshini.

Okubaluleke kakhulu, iningi lalezi zindlela zokusebenzisa zamanje lisahilela ukuhlelwa kwabantu kanye nokuphindaphinda . I-AI ingakhipha izithombe eziningi, futhi umuntu akhethe okungcono kakhulu futhi mhlawumbe ayithinte. Ngaleyo ndlela, i-AI isebenza ngokuzimela ukukhiqiza izinketho , kodwa abantu baqondisa isiqondiso sokudala futhi benza ukukhetha kokugcina. Ithembekile ekukhiqizeni okuqukethwe okuningi ngokushesha, kodwa ayiqinisekisiwe ukuhlangabezana nazo zonke izidingo ekuzameni kokuqala. Izinkinga ezifana nemininingwane engalungile (isb. i-AI idonsa izandla ngenombolo engalungile yeminwe, into engavamile eyaziwayo) noma imiphumela engahlosiwe isho ukuthi umqondisi wobuciko bomuntu ngokuvamile udinga ukuqapha ikhwalithi yokukhipha.

Nokho, kunezindawo lapho i-AI isondela khona ekuzimeleni okugcwele:

• Umklamo Okhiqizayo: Emasimini afana nokwakhiwa kwezakhiwo kanye nomklamo womkhiqizo, amathuluzi e-AI angakha ngokuzimela ama-prototypes okuklama ahlangabezana nemikhawulo ethile. Isibonelo, uma ubheka ubukhulu nemisebenzi efiselekayo yefenisha, i-algorithm yokukhiqiza ingaveza imiklamo eminingana esebenzayo (eminye engavamile) ngaphandle kokungenelela komuntu ngale kwemininingwane yokuqala. Le miklamo ingasetshenziswa ngokuqondile noma ihlanzwe ngabantu. Ngokufanayo, kobunjiniyela, i-AI yokukhiqiza ingaklama izingxenye (ake sithi, ingxenye yendiza) elungiselelwe isisindo namandla, ikhiqize izimo ezintsha umuntu angase angazicabanganga.

• Izimpahla Zemidlalo Yevidiyo: I-AI ingakhiqiza ukwakheka, amamodeli e-3D, noma ngisho namazinga aphelele emidlalo yevidiyo ngokuzenzakalelayo. Abathuthukisi basebenzisa lokhu ukusheshisa ukudalwa kokuqukethwe. Eminye imidlalo ye-indie isiqalile ukufaka ubuciko obukhiqizwe ngenqubo ngisho nengxoxo (ngamamodeli olimi) ukudala imihlaba yemidlalo emikhulu, enamandla enezimpahla ezincane ezenziwe ngabantu.

• I-Animation kanye nevidiyo (Okuvelayo): Nakuba ingavuthiwe njengezithombe ezimile, i-AI ekhiqizayo yevidiyo iyathuthuka. I-AI isivele ikwazi ukukhiqiza amavidiyo amafushane noma izithombe ezinyakazayo, yize ikhwalithi ingaguquki. Ubuchwepheshe be-Deepfake - obukhiqizayo - bungakhiqiza ukushintshana kobuso okungokoqobo noma ama-clone ezwi. Esimweni esilawulwayo, isitudiyo singasebenzisa i-AI ukukhiqiza isigcawu sangemuva noma izithombe ezinyakazayo zesixuku ngokuzenzakalelayo.

Okuphawulekayo ukuthi uGartner wabikezela ukuthi ngo-2030, sizobona ifilimu enkulu evelele enokuqukethwe okungu-90% okukhiqizwe yi-AI (kusukela kusikripthi kuya kokubonakalayo) ( I-Generative AI Use Cases for Industries and Enterprises ). Kusukela ngo-2025, asikafiki lapho - i-AI ayikwazi ukwenza ifilimu ende ngokuzimela. Kodwa izingcezu zaleyo puzzle ziyathuthuka: ukwenziwa kwesikripthi (umbhalo we-AI), ukwenziwa kwabalingiswa kanye nezigcawu (isithombe/ividiyo ye-AI), ukusebenza ngezwi (ama-clone ezwi le-AI), kanye nosizo lokuhlela (i-AI isingasiza kakade ngokusika nokushintsha).

Umbono we-2030-2035: Imidiya Ekhiqizwe yi-AI Esikalini

Uma sibheka phambili, indima ye-AI yokukhiqiza kwezobuciko bokubuka kanye nokuklama ilungele ukwanda kakhulu. Ngo-2035, silindele ukuthi i-AI izoba ngumdali wokuqukethwe oyinhloko emithonjeni eminingi yezindaba ezibukwayo, evame ukusebenza ngokufakwa okuncane kwabantu ngaphezu kwesiqondiso sokuqala. Okunye okulindelwe:

• Amafilimu Namavidiyo Akhiqizwe Ngokugcwele Nge-AI: Eminyakeni eyishumi ezayo, kungenzeka ukuthi sizobona ama-movie noma uchungechunge lokuqala olukhiqizwa kakhulu nge-AI. Abantu banganikeza isiqondiso sezinga eliphezulu (isb. uhlaka lwesikripthi noma isitayela esifunwayo) futhi i-AI izoveza izigcawu, idale ukufana kwabadlali, futhi igcizelele konke. Ukuhlolwa kokuqala kumafilimu amafushane kungenzeka kungakapheli iminyaka embalwa, ngemizamo yobude besici ngawo-2030. Lawa mafilimu e-AI angase aqale i-niche (i-animation yokuhlola, njll.) kodwa angaba yinto evamile njengoba ikhwalithi ithuthuka. Ukubikezela kwamafilimu kukaGartner okungu-90% ngo-2030 ( I-Generative AI Use Cases for Industries and Enterprises ), yize kunesifiso esikhulu, kugcizelela inkolelo yemboni yokuthi ukudalwa kokuqukethwe kwe-AI kuzoba yinkimbinkimbi ngokwanele ukuthwala umthwalo omkhulu ekwenzeni amafilimu.

• Ukuzenzakalela Kokuklama: Emikhakheni efana nemfashini noma izakhiwo, i-AI yokukhiqiza cishe izosetshenziselwa ukudweba ngokuzimela amakhulu emiqondo yokuklama ngokusekelwe kumapharamitha afana nokuthi "izindleko, izinto zokwakha, isitayela X", okushiya abantu bekhetha umklamo wokugcina. Lokhu kushintsha amandla amanje: esikhundleni sabaklami abakha kusukela ekuqaleni futhi mhlawumbe basebenzise i-AI ukuze baphefumulelwe, abaklami besikhathi esizayo bangase basebenze njengabagcini, bekhetha umklamo okhiqizwe yi-AI omuhle kakhulu futhi mhlawumbe bawulungise. Ngo-2035, umakhi angase afake izidingo zesakhiwo futhi athole amapulani aphelele njengeziphakamiso ezivela ku-AI (konke kuzwakala kahle ngokwesakhiwo, ngenxa yemithetho yobunjiniyela efakiwe).

• Ukudalwa Kokuqukethwe Okuqondene Nomuntu: Singase sibone ama-AI edala izithombe ezibukwayo ngokushesha kubasebenzisi ngabanye. Cabanga ngomdlalo wevidiyo noma okuhlangenwe nakho kweqiniso elingokoqobo ngo-2035 lapho indawo kanye nabalingiswa bevumelana nezintandokazi zomdlali, ezikhiqizwa ngesikhathi sangempela yi-AI. Noma imichilo yamahlaya eyenziwe ngokwezifiso ekhiqizwa ngokusekelwe osukwini lomsebenzisi - i-AI ezimele "yamahlaya yansuku zonke" eguqula ijenali yakho yombhalo ibe yimifanekiso ngokuzenzakalelayo njalo kusihlwa.

• Ukudala Okunezindlela Eziningi: Izinhlelo ze-AI ezikhiqizayo ziya ngokuya ziba nezindlela eziningi - okusho ukuthi zingakwazi ukuphatha umbhalo, izithombe, umsindo, njll. ndawonye. Ngokuhlanganisa lokhu, i-AI ingathatha isisho esilula njengokuthi “Ngenze umkhankaso wokumaketha womkhiqizo X” futhi ikhiqize hhayi nje ikhophi ebhaliwe, kodwa nezithombe ezihambisanayo, mhlawumbe ngisho neziqeshana zevidiyo ezimfushane zokuphromotha, konke kuhambisana ngesitayela. Lolu hlobo lwe-suite yokuqukethwe ngokuchofoza okukodwa luyisevisi engaba khona ekuqaleni kwawo-2030.

Ingabe i-AI izothatha indawo yabaculi abangabantu ? Lo mbuzo uvame ukuvela. Kungenzeka ukuthi i-AI izothatha umsebenzi omningi wokukhiqiza (ikakhulukazi ubuciko obuphindaphindwayo noma obushintsha ngokushesha obudingekayo ebhizinisini), kodwa ubuciko babantu buzohlala bukhona ngenxa yobuciko bokuqala kanye nokusungula izinto ezintsha. Ngo-2035, i-AI ezimele ingase idwebe isithombe ngendlela ethembekile ngesitayela somculi odumile - kodwa ukudala isitayela esisha noma ubuciko obuhambisana namasiko kungase kube yinto ebaluleke kakhulu kubantu (mhlawumbe nge-AI njengomhlanganyeli). Sibona ikusasa lapho abaculi abangabantu besebenza khona kanye "nabaculi abasebenzisana nabo" be-AI ezimele. Umuntu angathuma i-AI yomuntu siqu ukuthi ikhiqize ubuciko begalari yedijithali ekhaya lakhe, isibonelo, inikeze isimo sokudala esishintsha njalo.

Ngokombono wokuthembeka, i-AI ekhiqizayo ebonakalayo inendlela elula yokuzimela kunombhalo ngezindlela ezithile: isithombe singaba "sihle ngokwanele" ngokuzimela noma singaphelele, kanti iphutha eliyiqiniso embhalweni liyinkinga enkulu. Ngakho-ke, sesivele sibona ukwamukelwa okunengozi encane - uma umklamo owenziwe nge-AI mubi noma ungalungile, awuwusebenzisi nje, kodwa awuzilimazi ngokwawo. Lokhu kusho ukuthi ngawo-2030, izinkampani zingase zikhululeke ukuvumela i-AI ukuthi ikhiphe imiklamo ingagadiwe futhi ihilele abantu kuphela lapho kudingeka into entsha noma eyingozi ngempela.

Ngamafuphi, ngo-2035 i-AI ekhiqizayo kulindeleke ukuthi ibe ngumdali wokuqukethwe onamandla ezithombeni, okungenzeka ukuthi ibhekene nengxenye enkulu yezithombe nabezindaba ezisizungezile. Izokhiqiza okuqukethwe okuthembekile kwezokuzijabulisa, ukuklama, kanye nokuxhumana kwansuku zonke. Umdwebi ozimele useseduze - yize noma ngabe i-AI ibonwa njengokudala noma ithuluzi elihlakaniphile kakhulu kuyimpikiswano ezoshintsha njengoba imiphumela yayo ihluka kweyenziwe ngabantu.

I-AI Ekhiqizayo Ekuthuthukisweni Kwesofthiwe (Ukubhala Amakhodi)

Ukuthuthukiswa kwesofthiwe kungase kubonakale njengomsebenzi wokuhlaziya kakhulu, kodwa futhi kunesici sokudala - ukubhala ikhodi kuwukudala umbhalo ngolimi oluhlelekile. I-AI yokukhiqiza yesimanje, ikakhulukazi amamodeli ezilimi amakhulu, ibonakalise inekhono elikhulu ekubhaleni amakhodi. Amathuluzi afana ne-GitHub Copilot, i-Amazon CodeWhisperer, nabanye asebenza njengabahleli bezinhlelo ze-AI, ephakamisa iziqeshana zekhodi noma imisebenzi yonke njengoba abathuthukisi bezinhlobo. Lokhu kungafinyelela kude kangakanani ekuhleleni okuzenzakalelayo?

Amakhono Amanje (2025): I-AI njengoMqondisi Osizayo Wokubhala Amakhodi

Ngo-2025, abakhiqizi bekhodi ye-AI sebevame kakhulu emisebenzini eminingi yabathuthukisi. Lawa mathuluzi angaqedela ngokuzenzakalela imigqa yekhodi, akhiqize i-boilerplate (njengemisebenzi ejwayelekile noma izivivinyo), futhi abhale ngisho nezinhlelo ezilula ezinikezwe incazelo yolimi lwemvelo. Nokho, okubaluleke kakhulu, asebenza ngaphansi kokuqondisa komthuthukisi - umthuthukisi ubuyekeza futhi ahlanganise iziphakamiso ze-AI.

Amanye amaqiniso nezibalo zamanje:

• Abathuthukisi abangochwepheshe abangaphezu kwengxenye base bethathe abasizi bokubhala ikhodi ye-AI ngasekupheleni kuka-2023 ( Ukubhala ikhodi ku-Copilot: Idatha ka-2023 Iphakamisa Ukucindezeleka Okuphansi Ekhwalithini Yekhodi (kufaka phakathi ukubikezela kuka-2024) - GitClear ), okubonisa ukutholakala okusheshayo. I-GitHub Copilot, enye yamathuluzi okuqala atholakala kabanzi, kubikwe ukuthi ikhiqiza ngokwesilinganiso u-30-40% wekhodi kumaphrojekthi lapho isetshenziswa khona ( Ukubhala ikhodi akusekho i-MOAT. U-46% wamakhodi ku-GitHub usuvele ... ). Lokhu kusho ukuthi i-AI isivele ibhala izingxenye ezibalulekile zekhodi, yize umuntu eyiqondisa futhi eyiqinisekisa.

• Lawa mathuluzi e-AI ayazikhandla emisebenzini efana nokubhala ikhodi ephindaphindayo (isb., amakilasi emodeli yedatha, izindlela ze-getter/setter), ukuguqula ulimi olulodwa lokuhlela lube olunye, noma ukukhiqiza ama-algorithms aqondile afana nezibonelo zokuqeqesha. Isibonelo, unjiniyela angaphawula ngokuthi “// umsebenzi wokuhlunga uhlu lwabasebenzisi ngamagama” futhi i-AI izokhiqiza umsebenzi wokuhlunga ofanele cishe ngokushesha.

• Ziphinde zisize ekulungiseni amaphutha kanye nencazelo : abathuthukisi banganamathisela umlayezo wephutha bese i-AI ingase iphakamise ukulungiswa, noma ibuze ukuthi “Le khodi yenzani?” bese ithola incazelo yolimi lwemvelo. Lokhu kuyazimela ngandlela thile (i-AI ingazihlonza izinkinga ngokwayo), kodwa umuntu unquma ukuthi uzoyisebenzisa yini ukulungiswa.

• Okubalulekile, abasizi bamanje bokubhala amakhodi e-AI abawona amaphutha. Bangaphakamisa ikhodi engavikelekile, noma ikhodi ecishe ixazulule inkinga kodwa eneziphazamiso ezicashile. Ngakho-ke, umkhuba omuhle namuhla ukugcina umuntu enolwazi - unjiniyela uhlola futhi alungise ikhodi ebhalwe yi-AI njengoba nje ebengenza ikhodi ebhalwe ngabantu. Ezimbonini ezilawulwayo noma kwisofthiwe ebalulekile (njengezinhlelo zezokwelapha noma zezindiza), noma yimiphi iminikelo ye-AI ihlolwa ngokucophelela.

Akukho uhlelo lwesofthiwe olujwayelekile namuhla olusetshenziswa ngokuphelele yi-AI kusukela ekuqaleni ngaphandle kokubhekwa konjiniyela. Kodwa-ke, kuvela ezinye izindlela ezizimele noma ezizimele:

• Ukuhlolwa kweyunithi okukhiqizwa ngokuzenzakalela: I-AI ingahlaziya ikhodi futhi ikhiqize ukuhlolwa kweyunithi ukuze kuhlanganiswe amacala ahlukahlukene. Uhlaka lokuhlola lungakhiqiza futhi luqhube ngokuzenzakalelayo lokhu kuhlolwa okubhalwe yi-AI ukuze kubanjwe amaphutha, okwandisa ukuhlolwa okubhalwe ngabantu.

• Amapulatifomu anekhodi ephansi/angenakhodi ane-AI: Amanye amapulatifomu avumela abangebona abahleli bezinhlelo ukuthi bachaze lokho abakufunayo (isb. “bakha ikhasi lewebhu elinefomu lokuxhumana kanye nedathabheyisi ukuze balondoloze okufakiwe”) futhi uhlelo lukhiqiza ikhodi. Nakuba kusesezigabeni zokuqala, lokhu kubonisa ikusasa lapho i-AI ingadala khona isofthiwe ngokuzimela yamacala okusetshenziswa okujwayelekile.

• Ukubhala Ngesikripthi kanye Nekhodi Yeglue: Ukuzenzakalela kwe-IT kuvame ukuhilela ukubhala izikripthi ukuze kuxhunywe izinhlelo. Amathuluzi e-AI avame ukukhiqiza lezi zikripthi ezincane ngokuzenzakalelayo. Isibonelo, ukubhala iskripthi ukuze kuhlolwe ifayela lelogi bese kuthunyelwa isexwayiso se-imeyili - i-AI ingakhiqiza iskripthi esisebenzayo ngokuhlelwa okuncane noma okungenakho nhlobo.

Umbono we-2030-2035: Mayelana neSoftware “Ezithuthukisayo”

Eminyakeni eyishumi ezayo, i-AI ekhiqizayo kulindeleke ukuthi ithathe ingxenye enkulu yomthwalo wokubhala amakhodi, isondela ekuthuthukisweni kwesofthiwe ezimele ngokuphelele kwezigaba ezithile zamaphrojekthi. Ezinye zentuthuko ezilindelwe:

• Ukuqaliswa Okuphelele Kwesici: Ngo-2030, silindele ukuthi i-AI izokwazi ukusebenzisa izici ezilula zohlelo lokusebenza kusukela ekuqaleni kuya ekugcineni. Umphathi womkhiqizo angachaza isici ngolimi olulula ("Abasebenzisi kufanele bakwazi ukusetha kabusha iphasiwedi yabo ngesixhumanisi se-imeyili") futhi i-AI ingakhiqiza ikhodi edingekayo (ifomu elingaphambili, i-back-end logic, isibuyekezo sedathabheyisi, ukuthunyelwa kwe-imeyili) futhi iyihlanganise ku-codebase. I-AI ingasebenza kahle njengonjiniyela omncane ongalandela imininingwane. Unjiniyela ongumuntu angenza nje ukubuyekezwa kwekhodi futhi asebenzise izivivinyo. Njengoba ukuthembeka kwe-AI kuthuthuka, ukubuyekezwa kwekhodi kungaba ukuskena okusheshayo uma kunjalo.

• Ukugcinwa Kwekhodi Ezimele: Ingxenye enkulu yobunjiniyela besofthiwe akukhona nje ukubhala ikhodi entsha, kodwa nokubuyekeza ikhodi ekhona - ukulungisa amaphutha, ukuthuthukisa ukusebenza, ukuzivumelanisa nezidingo ezintsha. Abathuthukisi be-AI besikhathi esizayo cishe bazokwenza kahle kulokhu. Njengoba benikezwe i-codebase kanye nesiqondiso (“uhlelo lwethu lokusebenza luyawa lapho abasebenzisi abaningi kakhulu bengena ngesikhathi esisodwa”), i-AI ingase ithole inkinga (njengephutha le-concurrency) bese iyilungisa. Ngo-2035, izinhlelo ze-AI zingase ziphathe amathikithi okulungisa avamile ngokuzenzakalelayo ubusuku bonke, zisebenze njengeqembu lokulungisa elingapheli lezinhlelo zesofthiwe.

• Ukuhlanganiswa kanye nokusetshenziswa kwe-API: Njengoba izinhlelo eziningi zesofthiwe nama-API eza nemibhalo efundeka kalula, i-ejenti ye-AI ingathola ngokuzimela ukuthi ingaxhuma kanjani iSistimu A neNsizakalo B ngokubhala ikhodi yeglue. Isibonelo, uma inkampani ifuna uhlelo lwayo lwangaphakathi lwe-HR luvumelaniswe ne-API entsha yokukhokha, ingase inike i-AI umsebenzi "wokwenza lokhu kuxoxisane," futhi izobhala ikhodi yokuhlanganisa ngemva kokufunda imininingwane yazo zombili izinhlelo.

• Ikhwalithi Nokwenza Kahle: Amamodeli okukhiqiza ikhodi esikhathini esizayo cishe azofaka ama-feedback loops ukuqinisekisa ukuthi ikhodi iyasebenza (isb., ukusebenzisa izivivinyo noma ukulingisa ebhokisini lesanti). Lokhu kusho ukuthi i-AI ayikwazi ukubhala ikhodi kuphela kodwa futhi iyazilungisa ngokuyihlola. Ngo-2035, singacabanga nge-AI, uma inikezwe umsebenzi, eqhubeka nokuphindaphinda ikhodi yayo kuze kube yilapho zonke izivivinyo zidlula - inqubo umuntu angase angadingi ukuyiqapha umugqa nomugqa. Lokhu kuzokhulisa kakhulu ukwethemba ikhodi ekhiqizwe ngokuzimela.

Umuntu angabona isimo ngo-2035 lapho iphrojekthi encane yesofthiwe - ake sithi uhlelo lokusebenza lweselula olungokwezifiso lwebhizinisi - ingathuthukiswa kakhulu yi-ejenti ye-AI enikezwe imiyalelo ephezulu. "Unjiniyela" ongumuntu kuleso simo ungumphathi wephrojekthi noma umqinisekisi, ecacisa izidingo kanye nemingcele (ukuphepha, iziqondiso zesitayela) futhi evumela i-AI ukuthi yenze umsebenzi onzima wokubhala ikhodi yangempela.

Kodwa-ke, ngesofthiwe eyinkimbinkimbi, enkulu (izinhlelo zokusebenza, ama-algorithm e-AI athuthukile ngokwawo, njll.), ochwepheshe babantu basazobandakanyeka kakhulu. Ukuxazulula izinkinga zokudala kanye nomklamo wezakhiwo kusofthiwe kungenzeka kuhlale kuholwa ngabantu isikhathi esithile. I-AI ingase iphathe imisebenzi eminingi yokufaka amakhodi, kodwa ukunquma ukuthi yini ezokwakhiwa kanye nokuklama isakhiwo sonke kuyinselele ehlukile. Lokho kusho, njengoba i-AI ekhiqizayo iqala ukubambisana - ama-ejenti amaningi e-AI aphatha izingxenye ezahlukene zesistimu - kungenzeka ukuthi angaklama ngokubambisana izakhiwo ngezinga elithile (isibonelo, i-AI eyodwa iphakamisa ukwakheka kwesistimu, enye iyigxeke, futhi iyaphinda, nomuntu oqondisa inqubo).

Inzuzo enkulu elindelekile ye-AI ekubhaleni ikhodi ukukhulisa umkhiqizo . UGartner ubikezela ukuthi ngo-2028, u-90% wonjiniyela besofthiwe bazosebenzisa abasizi bekhodi ye-AI (kusuka ngaphansi kuka-15% ngo-2024) ( GitHub Copilot Tops Research Report on Assistants Assistants Akhodi ye-AI -- Visual Studio Magazine ). Lokhu kusikisela ukuthi abantu abangaphandle - labo abangasebenzisi i-AI - bazoba bambalwa. Singase sibone nokuntuleka kwabathuthukisi abangabantu ezindaweni ezithile ezincishiswa yi-AI egcwalisa izikhala; empeleni umthuthukisi ngamunye angenza okuningi kakhulu ngomsizi we-AI ongabhala ikhodi ngokuzimela.

Ukwethembana kuzohlala kuyindaba ebalulekile. Ngisho nango-2035, izinhlangano kuzodingeka ziqinisekise ukuthi ikhodi ekhiqizwe ngokuzimela iphephile (i-AI akumelwe ilethe ubuthakathaka) futhi ihambisana nemigomo yezomthetho/yokuziphatha (isb., i-AI ayifaki ikhodi ekopishiwe kusuka kulabhulali yomthombo ovulekile ngaphandle kwelayisensi efanele). Silindele amathuluzi okuphatha i-AI athuthukisiwe angaqinisekisa futhi alandelele imvelaphi yekhodi ebhalwe yi-AI ukusiza ukunika amandla ikhodi ezenzakalelayo ngaphandle kwengozi.

Ngamafuphi, maphakathi nawo-2030, i-AI ekhiqizayo cishe izosingatha isabelo esikhulu sokubhala amakhodi emisebenzini yesofthiwe evamile futhi isize kakhulu kweyinkimbinkimbi. Umjikelezo wokuphila kokuthuthukiswa kwesofthiwe uzoba ngokuzenzakalelayo kakhulu - kusukela ezidingweni kuya ekusetshenzisweni - kanye ne-AI engase ikhiqize futhi ithumele izinguquko zekhodi ngokuzenzakalelayo. Abathuthukisi abangabantu bazogxila kakhulu ekuqondeni okuseqophelweni eliphezulu, ulwazi lomsebenzisi, kanye nokuqapha, kuyilapho ama-ejenti e-AI ehlola imininingwane yokusetshenziswa.

I-AI Ekhiqizayo Ekusebenzeni Kwamakhasimende Nokusekela

Uma uke waxhumana nengxoxo yokusekelwa kwamakhasimende eku-inthanethi muva nje, kungenzeka ukuthi i-AI yayingakolunye uhlangothi okungenani ingxenye yayo. Isevisi yamakhasimende iyisizinda esilungele ukwenza ngokuzenzakalela kwe-AI: ihilela ukuphendula imibuzo yabasebenzisi, okuyinto i-AI ekhiqizayo (ikakhulukazi amamodeli engxoxo) engayenza kahle kakhulu, futhi ivame ukulandela izikripthi noma izihloko zolwazi, ezingafundwa yi-AI. I-AI ingaphatha kanjani amakhasimende ngokuzimela?

Amakhono Amanje (2025): Ama-Chatbot nama-Virtual Agents Athatha Umugqa Ongaphambili

Kusukela namuhla, izinhlangano eziningi zisebenzisa ama-chatbot e-AI njengendawo yokuqala yokuxhumana nesevisi yamakhasimende. Lawa asukela kuma-bot asekelwe emithethweni alula (“Cindezela u-1 ukuze uthole ukukhokhisa, u-2 ukuze uthole ukwesekwa…”) kuya kuma-chatbot e-AI athuthukile angachaza imibuzo ekhululekile futhi aphendule ngengxoxo. Amaphuzu abalulekile:

• Ukusingatha Imibuzo Evamile: Ama-ejenti e-AI ayazigqaja ekuphenduleni imibuzo evame ukubuzwa, ekuhlinzekeni ngolwazi (amahora esitolo, izinqubomgomo zokubuyiselwa kwemali, izinyathelo zokuxazulula izinkinga zezinkinga ezaziwayo), kanye nokuqondisa abasebenzisi ngezinqubo ezijwayelekile. Isibonelo, i-chatbot ye-AI yebhange ingasiza umsebenzisi ngokuzimela ukuthi ahlole ibhalansi ye-akhawunti yakhe, asethe kabusha iphasiwedi, noma achaze ukuthi angasifaka kanjani isicelo semali mboleko, ngaphandle kosizo lomuntu.

• Ukuqonda Ulimi Lwemvelo: Amamodeli anamuhla okukhiqiza avumela ukusebenzisana okuguquguqukayo nokufana "nokwabantu". Amakhasimende angathayipha umbuzo ngamazwi awo futhi i-AI ivame ukuqonda inhloso. Izinkampani zibika ukuthi ama-ejenti e-AI anamuhla aneliseka kakhulu kumakhasimende kunama-bot angaqondakali eminyaka embalwa edlule - cishe ingxenye yamakhasimende manje akholelwa ukuthi ama-ejenti e-AI angaba nozwela futhi asebenze kahle lapho ebhekana nezinkinga ( izibalo zesevisi yamakhasimende ze-AI ezingama-59 zango-2025 ), okubonisa ukwethembana okukhulayo enkonzweni eqhutshwa yi-AI.

• Ukusekelwa Kweziteshi Eziningi: I-AI ayigcini nje ngokuxoxa. Abasizi bezwi (njengezinhlelo ze-IVR zocingo ezine-AI ngemuva kwazo) sebeqala ukuphatha izingcingo, futhi i-AI ingabhala nezimpendulo ze-imeyili emibuzweni yamakhasimende ezingase ziphume ngokuzenzakalelayo uma kubhekwa ukuthi zinembile.

• Uma Abantu Bengena: Ngokuvamile, uma i-AI idideka noma umbuzo uyinkimbinkimbi kakhulu, uzodluliselwa kumenzeli womuntu. Izinhlelo zamanje ziyakwazi kahle ukwazi imikhawulo yazo ezimweni eziningi. Isibonelo, uma ikhasimende libuza into engavamile noma libonisa ukukhungatheka (“Lesi yisikhathi sesithathu ngikuthinta futhi ngicasukile kakhulu…”), i-AI ingase iphawule lokhu ukuze umuntu athathe izintambo. Umkhawulo wokudluliselwa ubekwa yizinkampani ukuze zilinganise ukusebenza kahle nokwaneliseka kwamakhasimende.

Izinkampani eziningi zibike izingxenye ezibalulekile zokusebenzisana ezixazululwa yi-AI yodwa. Ngokusho kocwaningo lwemboni, cishe u-70-80% wemibuzo yamakhasimende ejwayelekile ingaphathwa yi-chatbots ye-AI namuhla, kanti cishe u-40% wokusebenzisana kwamakhasimende ezinkampani kuzo zonke iziteshi sekuvele kuzenzakalelayo noma kusizakale nge-AI ( 52 Izibalo Zesevisi Yamakhasimende ze-AI Okufanele Uzazi - Plivo ). I-IBM's Global AI Adoption Index (2022) ibonise ukuthi u-80% wezinkampani zisebenzisa noma zihlela ukusebenzisa ama-chatbots e-AI ensizakaloni yamakhasimende ngo-2025.

Intuthuko ethokozisayo ukuthi i-AI ayigcini nje ngokuphendula amakhasimende, kodwa futhi isiza amanxusa abantu ngesikhathi sangempela. Isibonelo, ngesikhathi sengxoxo ebukhoma noma ucingo, i-AI ingase ilalele futhi inikeze inxusa labantu izimpendulo eziphakanyisiwe noma ulwazi olufanele ngokushesha. Lokhu kufiphaza umugqa wokuzimela - i-AI ayibhekene nekhasimende lodwa, kodwa ihileleke ngenkuthalo ngaphandle kombuzo ocacile womuntu. Isebenza kahle njengomeluleki ozimele wenxusa.

Ithemba lika-2030-2035: Ukusebenzisana kwamakhasimende okuqhutshwa kakhulu yi-AI

Ngo-2030, iningi lokuxhumana kwamakhasimende kulindeleke ukuthi libandakanye i-AI, kanti okuningi kuzophathwa ngokuphelele yi-AI kusukela ekuqaleni kuze kube sekupheleni. Izibikezelo kanye nezitayela ezisekela lokhu:

• Imibuzo Eyinkimbinkimbi Ephakeme Ixazululiwe: Njengoba amamodeli e-AI ehlanganisa ulwazi olubanzi futhi ethuthukisa ukucabanga, azokwazi ukusingatha izicelo zamakhasimende eziyinkimbinkimbi kakhulu. Esikhundleni sokuphendula nje ngokuthi “Ngiyibuyisela kanjani into?”, i-AI yesikhathi esizayo ingase ibhekane nezinkinga eziningi ezifana nokuthi, “I-inthanethi yami ayisebenzi, ngizamile ukuqala kabusha, ungasiza?” ngokuxilonga inkinga ngengxoxo, ukuqondisa ikhasimende ekuxazululeni izinkinga okuthuthukisiwe, futhi kuphela uma konke okunye kwehluleka ukuhlela uchwepheshe - imisebenzi namuhla engase idinge ubuchwepheshe bokusekela abantu. Enkonzweni yamakhasimende ezempilo, i-AI ingase ibhekane nokuhlela ama-aphoyintimenti esiguli noma imibuzo yomshuwalense kusukela ekuqaleni kuze kube sekupheleni.

• Isixazululo Sesevisi Ephela: Singase sibone i-AI ingagcini nje ngokutshela ikhasimende ukuthi lenzeni, kodwa empeleni ikwenze egameni lekhasimende ngaphakathi kwezinhlelo ze-backend. Isibonelo, uma ikhasimende lithi “Ngifuna ukushintsha indiza yami ibe ngoMsombuluko ozayo bese ngengeza esinye isikhwama,” i-ejenti ye-AI ngo-2030 ingase ixhumane ngqo nesistimu yokubhukha yenkampani yezindiza, yenze ushintsho, icubungule inkokhelo yesikhwama, futhi iqinisekise kumakhasimende - konke ngokuzimela. I-AI iba yi-ejenti yesevisi ephelele, hhayi nje umthombo wolwazi.

• Ama-Agent e-AI Akhona Yonke Indawo: Izinkampani cishe zizosebenzisa i-AI kuzo zonke izindawo zokuxhumana namakhasimende - ngocingo, ngengxoxo, nge-imeyili, nakwezokuxhumana. Amakhasimende amaningi angase angaqapheli ukuthi akhuluma ne-AI noma umuntu, ikakhulukazi njengoba amazwi e-AI eba ngokwemvelo futhi izimpendulo zengxoxo ziqaphela kakhulu umongo. Ngo-2035, ukuxhumana nesevisi yamakhasimende kungasho ukuxhumana ne-AI ehlakaniphile ekhumbula ukusebenzisana kwakho kwangaphambilini, eqonda okuthandayo, futhi evumelana nethoni yakho - empeleni i-ejenti ebonakalayo eyenzelwe wena wonke amakhasimende.

• Ukwenza Izinqumo ze-AI Ekusebenzisaneni: Ngaphandle kokuphendula imibuzo, i-AI izoqala ukwenza izinqumo ezidinga imvume yokuphatha njengamanje. Isibonelo, namuhla i-ejenti yomuntu ingadinga imvume yomphathi ukuze inikeze imbuyiselo noma isaphulelo esikhethekile ukuze ithokozise ikhasimende elithukuthele. Esikhathini esizayo, i-AI ingaphathiswa lezo zinqumo, ngaphakathi kwemingcele echaziwe, ngokusekelwe enanini elibalwe lempilo yamakhasimende kanye nokuhlaziywa kwemizwa. Ucwaningo olwenziwe yi-Futurum/IBM lubikezele ukuthi ngo-2030 cishe u-69% wezinqumo ezenziwe ngesikhathi sangempela sokuzibandakanya kwamakhasimende zizokwenziwa yimishini ehlakaniphile (Ukucabanga Kabusha Ngokushintsha Ushintsho Ku-CX, Abamakethi Kumelwe Bazenze Lezi Zinto Ezi-2 ) - ngempumelelo i-AI inquma inkambo engcono kakhulu yesenzo ekusebenzisaneni.

• Ukubandakanyeka kwe-AI okungu-100%: Umbiko owodwa uphakamisa ukuthi i-AI izogcina idlale indima kukho konke ukuxhumana kwamakhasimende ( izibalo zesevisi yamakhasimende ze-AI ezingama-59 zika-2025 ), kungaba kusenesikhathi noma ngemuva. Lokho kungasho ukuthi noma umuntu exhumana nekhasimende, uzosizwa yi-AI (ukunikeza iziphakamiso, ukuthola ulwazi). Ngaphandle kwalokho, incazelo iwukuthi akukho mbuzo wamakhasimende ongaphendulwa nganoma yisiphi isikhathi - uma abantu bengaxhunyiwe ku-inthanethi, i-AI ihlala ikhona.

Ngo-2035, singase sithole ukuthi ama-ejenti esevisi yamakhasimende abantu asevele egxile kuphela ezimweni ezibucayi kakhulu noma ezithinta kakhulu (isib., amaklayenti e-VIP noma isixazululo sezikhalazo esiyinkimbinkimbi esidinga uzwela lomuntu). Imibuzo evamile - kusukela ebhange kuya ekuthengiseni kuya ekusekelweni kobuchwepheshe - ingahlinzekwa yiqembu lama-ejenti e-AI asebenza amahora angama-24 ngosuku, izinsuku eziyi-7 ngesonto, efunda njalo ekusebenzisaneni ngakunye. Lolu shintsho lungenza isevisi yamakhasimende ihambisane futhi isheshe, njengoba i-AI ingabagcini abantu belindile futhi ngokombono ingenza imisebenzi eminingi ukuze iphathe amakhasimende angenamkhawulo ngasikhathi sinye.

Kunezinselele okufanele zinqotshwe kulo mbono: I-AI kumele ibe namandla kakhulu ukuze ibhekane nokungalindeleki kwamakhasimende angabantu. Kumele ikwazi ukubhekana ne-slang, intukuthelo, ukudideka, kanye nezindlela eziningi ezingapheli zokuxhumana nabantu. Idinga futhi ulwazi olusha (akunandaba ukuthi ulwazi lwe-AI seluphelelwe yisikhathi). Ngokutshala imali ekuhlanganisweni phakathi kwe-AI kanye nezizindalwazi zenkampani (ulwazi lwesikhathi sangempela mayelana nama-oda, ukungasebenzi, njll.), lezi zithiyo zingaxazululwa.

Ngokwezimiso zokuziphatha, izinkampani kuzodingeka zinqume ukuthi zizodalula nini ukuthi “ukhuluma ne-AI” futhi ziqinisekise ubulungiswa (i-AI ayiphathi amakhasimende athile ngendlela engeyinhle ngenxa yokuqeqeshwa okunobandlululo). Uma lokhu kuphathwa, icala lebhizinisi liqinile: Insizakalo yamakhasimende ye-AI inganciphisa kakhulu izindleko nezikhathi zokulinda. Imakethe ye-AI enkonzweni yamakhasimende kulindeleke ukuthi ikhule ibe ngamashumi ezigidigidi zamaRandi ngo-2030 (I- AI kuMbiko Wemakethe Yensizakalo Yamakhasimende 2025-2030: Icala ) ( Indlela i-AI Ekhiqiza Ngayo Ithuthukisa Ukuphathwa Kwempahla | Ryder ) njengoba izinhlangano zitshala imali kula makhono.

Ngamafuphi, lindela ikusasa lapho isevisi yamakhasimende e-AI ezimele iyinto evamile . Ukuthola usizo ngokuvamile kusho ukusebenzisana nomshini ohlakaniphile ongaxazulula inkinga yakho ngokushesha. Abantu bazobe besesesikhathini sokuqapha nokusingatha amacala abucayi, kodwa kakhulu njengabaqondisi bezisebenzi ze-AI. Umphumela ungaba isevisi esheshayo, eyenzelwe wena kubathengi - uma nje i-AI iqeqeshwe kahle futhi iqashwe ukuze kuvinjelwe ukukhungatheka kokuhlangenwe nakho "kwe-robot hotline" kwesikhathi esidlule.

I-AI Ekhiqizayo Kwezempilo Nezokwelapha

Ukunakekelwa kwezempilo kuyinsimu lapho izinkinga ziphezulu khona. Umqondo wokuthi i-AI isebenza ngaphandle kokubhekwa ngumuntu kwezokwelapha uvusa injabulo (yokusebenza kahle kanye nokufinyelela) kanye nokuqapha (ngezizathu zokuphepha kanye nozwela). I-AI ekhiqizayo isiqalile ukungena ezindaweni ezifana nokuhlaziywa kwezithombe zezokwelapha, imibhalo yezokwelapha, ngisho nokutholakala kwemithi. Yini engayenza ngokwayo ngokuzibophezela?

Amakhono Amanje (2025): Ukusiza Odokotela, Hhayi Ukubashintsha

Okwamanje, i-AI ekhiqizayo kwezempilo isebenza kakhulu njengomsizi onamandla kochwepheshe bezokwelapha, kunokuba ngumenzi wezinqumo ozimele. Isibonelo:

• Imibhalo Yezokwelapha: Enye yezindlela eziphumelela kakhulu zokusebenzisa i-AI ekunakekelweni kwezempilo ukusiza odokotela ngamaphepha. Amamodeli olimi lwemvelo angabhala ukuvakashelwa kweziguli futhi akhiqize amanothi emtholampilo noma izifinyezo zokukhishwa. Izinkampani zine-"AI scribers" ezilalela ngesikhathi sokuhlolwa (ngemakrofoni) futhi zikhiqize ngokuzenzakalelayo uhlaka lwamanothi okuhlangana ukuze udokotela abukeze. Lokhu kusindisa odokotela isikhathi sokuthayipha. Ezinye izinhlelo zize zigcwalise izingxenye zamarekhodi ezempilo kagesi. Lokhu kungenziwa ngokungenelela okuncane - udokotela ulungisa noma yimaphi amaphutha amancane kuhlaka, okusho ukuthi ukubhala amanothi kuzimele kakhulu.

• I-Radiology kanye ne-Imaging: I-AI, okuhlanganisa namamodeli okukhiqiza, ingahlaziya ama-X-ray, ama-MRI, kanye nama-CT scan ukuthola ama-anomaly (njengezimila noma ukwaphuka). Ngo-2018, i-FDA yavuma uhlelo lwe-AI lokuthola ngokuzimela i-retinopathy yesifo sikashukela (isimo samehlo) ezithombeni ze-retina - ikakhulukazi, yagunyazwa ukwenza ucingo ngaphandle kokubuyekezwa kochwepheshe kulowo mongo wokuhlola othize. Leso simiso sasingeyona i-AI yokukhiqiza, kodwa sibonisa ukuthi abalawuli bavumele ukuxilongwa kwe-AI yokuzimela ezimweni ezilinganiselwe. Amamodeli okukhiqiza ayasebenza ekudaleni imibiko ephelele. Isibonelo, i-AI ingase ihlole i-X-ray yesifuba bese ibhala umbiko we-radiologist ethi "Akukho okutholakele okuphuthumayo. Amaphaphu acacile. Usayizi ojwayelekile wenhliziyo." I-radiologist ibe isiqinisekisa futhi iyisayine. Kwezinye izimo ezivamile, le mibiko ingaphuma ngaphandle kokuhlelwa uma i-radiologist ithembela i-AI futhi yenza nje ukuhlola okusheshayo.

• Abahloli Bezimpawu kanye Nabahlengikazi Abasebenzisa I-Virtual: Ama-chatbot e-AI akhiqizayo asetshenziswa njengabahloli bezimpawu abaphambili. Iziguli zingafaka izimpawu zazo futhi zithole iseluleko (isb., “Kungaba umkhuhlane ovamile; ukuphumula kanye noketshezi, kodwa bona udokotela uma kuvela u-X noma u-Y.”). Izinhlelo zokusebenza ezifana ne-Babylon Health zisebenzisa i-AI ukunikeza izincomo. Njengamanje, lezi zivame ukuhlelwa njengeseluleko solwazi, hhayi iseluleko sezokwelapha esiqondile, futhi zikhuthaza ukulandelwa nodokotela wabantu ngezinkinga ezinkulu.

• Ukutholakala Kwezidakamizwa (Ikhemistri Ekhiqizayo): Amamodeli e-AI akhiqizayo angaphakamisa izakhiwo ezintsha zama-molecule emithi. Lokhu kusemkhakheni wocwaningo kunokunakekelwa kweziguli. Lawa ma-AI asebenza ngokuzimela ukuze aphakamise izinkulungwane zamakhemikhali afanelekayo anezakhiwo ezifunwayo, ososayensi bamakhemikhali babantu abawabuyekezayo bese bewahlola elabhorethri. Izinkampani ezifana ne-Insilico Medicine zisebenzise i-AI ukukhiqiza izidakamizwa ezintsha ngesikhathi esincane kakhulu. Nakuba lokhu kungaxhumani ngqo neziguli, kuyisibonelo se-AI edala ngokuzimela izixazululo (imiklamo yama-molecule) abantu ababengathatha isikhathi eside ukuzithola.

• Imisebenzi Yezempilo: I-AI isiza ukuthuthukisa ukuhlela, ukuphathwa kokuhlinzekwa, kanye nokunye ukuthutha ezibhedlela. Isibonelo, imodeli yokukhiqiza ingase ilingise ukugeleza kweziguli futhi iphakamise ukulungiswa kokuhlela ukuze kuncishiswe izikhathi zokulinda. Nakuba kungabonakali kangako, lezi yizinqumo i-AI engayenza ngezinguquko ezincane ezenziwe ngesandla.

Kubalulekile ukusho ukuthi kusukela ngo-2025, asikho isibhedlela esivumela i-AI ukuthi yenze izinqumo ezinkulu zezokwelapha noma ukwelashwa ngaphandle kokuvunyelwa ngumuntu. Ukuhlonza kanye nokuhlela ukwelashwa kusesezandleni zabantu, kanti i-AI inikeza umbono. Ukwethenjwa okudingekayo ukuze i-AI itshele isiguli ngokuzimela ukuthi “Unomdlavuza” noma ukunikeza imithi akukabi khona okwamanje, futhi akufanele kube ngaphandle kokuqinisekiswa okubanzi. Ochwepheshe bezokwelapha basebenzisa i-AI njengesibili samehlo noma njengethuluzi lokonga isikhathi, kodwa baqinisekisa imiphumela ebalulekile.

Umbono ka-2030-2035: I-AI njengozakwabo kadokotela (futhi mhlawumbe nomhlengikazi noma usokhemisi)

Eminyakeni eyishumi ezayo, silindele ukuthi i-AI yokukhiqiza ithathe imisebenzi ejwayelekile yezokwelapha ngokuzimela futhi ithuthukise ukufinyelela kwezinsizakalo zezempilo:

• Ukuxilongwa Kwangaphambi Kokuqala Okuzenzakalelayo: Ngo-2030, i-AI yayingaphatha ngokwethembeka ukuhlaziywa kokuqala kwezimo eziningi ezivamile. Cabanga ngohlelo lwe-AI emtholampilo olufunda izimpawu zesiguli, umlando wezokwelapha, ngisho nethoni yaso kanye nezimpawu zobuso ngekhamera, futhi lunikeze isiphakamiso sokuxilonga kanye nokuhlolwa okunconywayo - konke lokhu ngaphambi kokuba udokotela womuntu abone isiguli. Udokotela angagxila ekuqinisekiseni nasekuxoxeni ngokuxilongwa. Kwezokwelapha nge-telemedicine, isiguli singase siqale sixoxe ne-AI enciphisa inkinga (isb., ukutheleleka okungenzeka kwe-sinus vs. okuthile okunzima kakhulu) bese sibaxhumanisa nodokotela uma kudingeka. Abalawuli bangavumela i-AI ukuthi ngokusemthethweni izimo ezithile ezincane ngaphandle kokubhekwa ngumuntu uma kutholakale ukuthi kunembile kakhulu - isibonelo, i-AI ihlola ukutheleleka kwendlebe okuqondile kusuka esithombeni se-otoscope kungenzeka.

• Abaqaphi Bempilo Yomuntu Siqu: Njengoba kwanda izinto ezigqokwayo (amawashi ahlakaniphile, izinzwa zezempilo), i-AI izoqapha iziguli njalo futhi izixwayise ngokuzimela ngezinkinga. Isibonelo, ngo-2035 i-AI yakho egqokwayo ingase ibone isigqi senhliziyo esingajwayelekile futhi ikuhlelele ngokuzimela ukubonisana okuphuthumayo nge-inthanethi noma ngisho nokushayela i-ambulensi uma ithola izimpawu zokuhlaselwa yinhliziyo noma isifo sohlangothi. Lokhu kudlulela endaweni yokuzikhethela ngokuzimela - ukunquma ukuthi isimo siphuthumayo futhi siyasebenza - okuwukusetshenziswa kwe-AI okungenzeka futhi okusindisa impilo.

• Izincomo Zokwelashwa: I-AI ekhiqizayo eqeqeshwe ngezincwadi zezokwelapha kanye nedatha yesiguli ingase iphakamise izinhlelo zokwelapha ezenzelwe wena. Ngo-2030, ngezifo eziyinkimbinkimbi njengomdlavuza, amabhodi esimila se-AI angahlaziya ukwakheka kwezakhi zofuzo zesiguli kanye nomlando wezokwelapha futhi abhale ngokuzimela uhlelo lokwelashwa olunconywayo (uhlelo lwe-chemo, ukukhethwa kwemithi). Odokotela babantu bazolubuyekeza, kodwa ngokuhamba kwesikhathi njengoba ukuzethemba kwakheka, bangase baqale ukwamukela izinhlelo ezikhiqizwe yi-AI ikakhulukazi ezimweni ezivamile, balungise kuphela lapho kudingeka.

• Abahlengikazi Abasebenzisa I-inthanethi Nokunakekelwa Kwasekhaya: I-AI engakwazi ukuxoxa futhi inikeze isiqondiso sezokwelapha ingaphatha ukulandelela okuningi kanye nokuqapha ukunakekelwa okungapheli. Isibonelo, iziguli ezisekhaya ezinezifo ezingapheli zingabika izibalo zansuku zonke kumsizi womhlengikazi we-AI onikeza iseluleko (“Ushukela wakho wegazi uphezulu kancane, cabanga ukulungisa ukudla kwakho kwakusihlwa”) bese kuthi kuphela umhlengikazi ongumuntu angene lapho ukufundwa kungaphandle kwebanga noma kuvela izinkinga. Le-AI ingasebenza ngokuzimela ngaphansi kokuqashwa kude kukadokotela.

• Ukuhlaziywa Kwezithombe Zezokwelapha Nokuhlolwa Kwelebhu – Amapayipi Azenzakalelayo Ngokugcwele: Ngo-2035, ukufunda ukuskena kwezokwelapha kungenzeka kwenziwe kakhulu yi-AI kweminye imikhakha. Ochwepheshe bemisebe bazoqondisa izinhlelo ze-AI futhi baphathe amacala ayinkimbinkimbi, kodwa iningi lokuskena okuvamile (okuyinto evamile ngempela) "kungafundwa" futhi kusayinwe yi-AI ngqo. Ngokufanayo, ukuhlaziya amaslayidi e-pathology (isibonelo, ukuthola amangqamuzana omdlavuza ku-biopsy) kungenziwa ngokuzimela ukuze kuhlolwe okokuqala, kusheshiswe kakhulu imiphumela yelebhu.

• Ukutholakala Kwezidakamizwa Nezilingo Zemitholampilo: I-AI cishe ngeke iklame ama-molecule ezidakamizwa kuphela kodwa futhi izokhiqiza idatha yeziguli zokwenziwa zezilingo noma ithole abantu abazongenela ukuhlolwa okufanele. Ingase isebenzise ngokuzimela izilingo ezingokoqobo (ilingise indlela iziguli ezingasabela ngayo) ukuze inciphise izinketho ngaphambi kwezilingo zangempela. Lokhu kungaletha imithi emakethe ngokushesha ngezilingo ezimbalwa eziqhutshwa ngabantu.

Umbono kadokotela we-AI ozongena esikhundleni sikadokotela ongumuntu ngokuphelele usekude kakhulu futhi usalokhu uphikisana. Ngisho nango-2035, kulindeleke ukuthi i-AI izosebenza njengomlingani wodokotela esikhundleni sokushintsha ukuthinta komuntu. Ukuxilongwa okuyinkimbinkimbi kuvame ukudinga umuzwa, ukuziphatha, kanye nezingxoxo ukuze kuqondwe umongo wesiguli - izindawo lapho odokotela abangabantu bephumelela khona. Lokho kusho ukuthi, i-AI ingase iphathe, ake sithi, u-80% womsebenzi ojwayelekile: amaphepha, amacala aqondile, ukuqapha, njll., okuvumela odokotela abangabantu ukuthi bagxile ku-20% onzima kanye nobudlelwano besiguli.

Kunezingqinamba ezibalulekile: ukuvunyelwa komthetho kwe-AI ezimele ekunakekelweni kwezempilo kunzima (kufanele kanjalo). Izinhlelo ze-AI zizodinga ukuqinisekiswa okubanzi kwezokwelapha. Singase sibone ukwamukelwa okuqhubekayo - isib., i-AI ivunyelwe ukuxilonga noma ukwelapha ngokuzimela ezindaweni ezinganakekelwa kahle lapho kungekho odokotela abatholakala khona, njengendlela yokwandisa ukufinyelela kwezempilo (cabanga "ngomtholampilo we-AI" edolobhaneni elikude ngo-2030 osebenza ngokuqondiswa okwenziwa ngocingo njalo ngudokotela edolobheni).

Izinto ezicatshangelwayo zokuziphatha zibalulekile. Ukuzibophezela (uma i-AI ezimele yenza iphutha ekuxilongweni, ubani onesibopho?), imvume enolwazi (iziguli zidinga ukwazi ukuthi i-AI ihilelekile yini ekunakekelweni kwazo), kanye nokuqinisekisa ukulingana (i-AI isebenza kahle kubo bonke abantu, igwema ukucwasa) kuyinselele okufanele zihambe kuyo. Uma lokho sekuxazululiwe, maphakathi nawo-2030 i-AI ekhiqizayo ingase ihlanganiswe nokulethwa kwezempilo, yenze imisebenzi eminingi ekhulula abahlinzeki babantu futhi ifinyelele iziguli ezinokufinyelela okulinganiselwe okwamanje.

Ngamafuphi, ngo-2035 ukunakekelwa kwezempilo cishe kuzobona i-AI ihlanganiswe ngokujulile kodwa ikakhulukazi ngaphansi kwesibopho noma ezindimeni zokusekela. Sizothemba i-AI ukuthi yenze okuningi ngokwayo - ukufunda amaskeni, ukubuka izinto ezibalulekile, izinhlelo zokubhala - kodwa inetha lokuphepha lokuqapha kwabantu lisesekhona ukuze kwenziwe izinqumo ezibalulekile. Umphumela ungaba uhlelo lwezempilo olusebenza kahle nolusabelayo, lapho i-AI iphatha khona ukuthwala okunzima futhi abantu banikeza uzwela kanye nesahlulelo sokugcina.

I-AI Ekhiqizayo Kwezemfundo

Imfundo ingenye insimu lapho i-AI ekhiqizayo yenza khona amagagasi, kusukela kuma-bots okufundisa asebenzisa i-AI kuya kuma-grade okuzenzakalelayo kanye nokudalwa kokuqukethwe. Ukufundisa nokufunda kuhilela ukuxhumana kanye nokudala, okuyizikhwepha zamamodeli okukhiqiza. Kodwa ingabe i-AI ingathenjwa ukuthi ifundise ngaphandle kokugadwa nguthisha?

Amakhono Amanje (2025): Abafundisi kanye Nabakhiqizi Bokuqukethwe Ngentambo

Njengamanje, i-AI isetshenziswa kwezemfundo ikakhulukazi njengethuluzi elingeziwe kunokuba ibe uthisha ozimele. Izibonelo zokusetshenziswa kwamanje:

• Abasizi Bokufundisa Nge-AI: Amathuluzi afana ne-“Khanmigo” yeKhan Academy (enikwe amandla yi-GPT-4) noma izinhlelo zokusebenza ezahlukene zokufunda ulimi zisebenzisa i-AI ukulingisa umfundisi womuntu nomuntu noma umlingani wengxoxo. Abafundi bangabuza imibuzo ngolimi lwemvelo futhi bathole izimpendulo noma izincazelo. I-AI inganikeza amacebiso ngezinkinga zomsebenzi wasekhaya, ichaze imiqondo ngezindlela ezahlukene, noma idlale indima njengomuntu odumile wesifundo somlando esisebenzisanayo. Kodwa-ke, laba bafundisi be-AI bavame ukusetshenziswa ngokunganaki; othisha noma abanakekeli bohlelo lokusebenza bavame ukuqapha izingxoxo noma babeke imingcele kulokho i-AI engaxoxa ngakho (ukugwema ulwazi olungelona iqiniso noma okuqukethwe okungafanele).

• Ukudalwa Kokuqukethwe Kothisha: I-AI Ekhiqizayo isiza othisha ngokudala imibuzo yemibuzo, izifinyezo zokufunda, izinhlaka zohlelo lwezifundo, njalo njalo. Uthisha angase abuze i-AI, “Khiqiza izinkinga zokuzijwayeza ezi-5 kuma-quadratic equations anezimpendulo,” onge isikhathi sokulungiselela. Lokhu ukukhiqiza okuqukethwe okuzenzakalelayo, kodwa uthisha uvame ukubukeza umphumela ukuze athole ukunemba nokuhambisana nekharikhulamu. Ngakho-ke kuyithuluzi elonga abasebenzi kakhulu kunelizimele ngokuphelele.

• Ukuthola Amamaki kanye Nempendulo: I-AI ingakwazi ukuphawula ngokuzenzakalelayo izivivinyo zokukhetha okuningi (akukho okusha lapho) futhi ingahlaziya izimpendulo ezimfushane noma izindatshana. Ezinye izinhlelo zesikole zisebenzisa i-AI ukuphawula izimpendulo ezibhaliwe futhi zinikeze impendulo kubafundi (isb., ukulungiswa kohlelo lolimi, iziphakamiso zokwandisa impikiswano). Nakuba kungewona umsebenzi wokukhiqiza ngokwawo, ama-AI amasha angakhiqiza ngisho nombiko wempendulo owenzelwe umfundi ngokusekelwe ekusebenzeni kwakhe, eqokomisa izindawo okufanele zithuthukiswe. Othisha bavame ukuhlola kabili izindatshana ezifakwe amamaki e-AI kulesi sigaba ngenxa yokukhathazeka mayelana nokungafani.

• Izinhlelo Zokufunda Ezivumelana Nezimo: Lezi yizinkundla ezilungisa ubunzima noma isitayela sezinto ezisekelwe ekusebenzeni komfundi. I-AI Ekhiqizayo ithuthukisa lokhu ngokudala izinkinga ezintsha noma izibonelo ezilungiselelwe izidingo zomfundi ngokushesha. Isibonelo, uma umfundi enenkinga ngomqondo, i-AI ingase idale omunye umbuzo wokufanisa noma wokuzijwayeza ogxile kulowo mqondo. Lokhu kuzimele ngandlela thile, kodwa ngaphakathi kohlelo oluklanywe othisha.

• Ukusetshenziswa Kwabafundi Ekufundeni: Abafundi ngokwabo basebenzisa amathuluzi afana ne-ChatGPT ukusiza ekufundeni - ukucela ukucaciswa, ukuhumusha, noma ngisho nokusebenzisa i-AI ukuthola impendulo ngohlaka lwendatshana (“ukuthuthukisa isigaba sami sesingeniso”). Lokhu kuqondiswa bona futhi kungaba ngaphandle kolwazi lothisha. I-AI kulesi simo isebenza njengomfundisi noma umlungisi wokuhlola uma kudingeka. Inselele ukuqinisekisa ukuthi abafundi bayisebenzisela ukufunda kunokuthola nje izimpendulo (ubuqotho bezemfundo).

Kusobala ukuthi kusukela ngo-2025, i-AI kwezemfundo inamandla kodwa ngokuvamile isebenza nomfundisi ongumuntu oqondisa iminikelo ye-AI. Kunesixwayiso esiqondakalayo: asifuni ukwethemba i-AI ukuthi izofundisa ulwazi olungalungile noma iphathe ukuxhumana kwabafundi okubucayi endaweni engenamuntu. Othisha babheka abafundisi be-AI njengabasizi abawusizo abanganikeza abafundi ukuzijwayeza okwengeziwe nezimpendulo ezisheshayo zemibuzo evamile, okukhulula othisha ukuthi bagxile ekuqeqeshweni okujulile.

Umbono ka-2030-2035: Abafundisi be-AI abazimele kanye nabasizi bokufundisa abazenzakalelayo

Eminyakeni eyishumi ezayo, silindele ukuthi i-AI ekhiqizayo izokwenza kube nokwenzeka ukufunda okuqondene nomuntu siqu nokuzimela , kuyilapho izindima zothisha zithuthuka:

• Abafundisi Bomuntu Siqu be-AI Bawo Wonke Umfundi: Ngo-2030, umbono (owabiwe ochwepheshe abanjengoSal Khan waseKhan Academy) ukuthi umfundi ngamunye angakwazi ukufinyelela kumfundisi we-AI osebenza kahle njengomfundisi ongumuntu ngezindlela eziningi ( Lo mfundisi we-AI angenza abantu bahlakaniphe izikhathi eziyi-10, kusho umdali wawo ). Laba bafundisi be-AI bazotholakala amahora angama-24 ngosuku, izinsuku eziyi-7 ngesonto, bazi umlando wokufunda womfundi ngokuseduze, futhi bavumelanise isitayela sabo sokufundisa ngokufanele. Isibonelo, uma umfundi engumfundi obonakalayo ohlushwa umqondo we-algebra, i-AI ingase idale incazelo ebonakalayo noma ukulingisa okusebenzisanayo ukusiza. Ngenxa yokuthi i-AI ingalandelela intuthuko yomfundi ngokuhamba kwesikhathi, inganquma ngokuzimela ukuthi yisiphi isihloko okufanele isibukeze ngokulandelayo noma ukuthi izothuthukela nini ekhonweni elisha - ukuphatha ngempumelelo uhlelo lwesifundo salowo mfundi ngomqondo omncane.

• Umthwalo Womsebenzi Wothisha Oncishisiwe Emisebenzini Ejwayelekile: Ukufaka amamaki, ukwenza amaphepha okusebenzela, ukubhala phansi izinto zesifundo – le misebenzi ingase ikhishwe ngokuphelele ku-AI ngawo-2030. I-AI ingakhiqiza umsebenzi wesikole owenziwe ngokwezifiso wesonto lonke wekilasi, ibeke amamaki kuzo zonke izabelo zesonto eledlule (ngisho nezivulekile) ngempendulo, futhi igqamise kuthisha ukuthi yibaphi abafundi abangadinga usizo olwengeziwe ngezihloko. Lokhu kungenzeka ngokufakwa okuncane kothisha, mhlawumbe ukubuka nje okusheshayo ukuqinisekisa ukuthi amamaki e-AI abonakala efanele.

• Amapulatifomu Okufunda Okuzivumelanisa Nezimo: Singase sibone izifundo eziqhutshwa yi-AI ngokugcwele zezifundo ezithile. Cabanga ngesifundo esiku-inthanethi esingenaye umfundisi ongumuntu lapho i-ejenti ye-AI yethula khona izinto, inikeze izibonelo, iphendula imibuzo, futhi ilungisa ijubane ngokusekelwe kumfundi. Okuhlangenwe nakho komfundi kungaba okuhlukile kuye, okudalwe ngesikhathi sangempela. Ukuqeqeshwa kwenkampani kanye nokufunda kwabantu abadala kungase kuthuthele kule modeli maduze, lapho ngo-2035 isisebenzi singathi “Ngifuna ukufunda ama-macro e-Excel athuthukile” bese umfundisi we-AI ezobafundisa ngekharikhulamu eyenzelwe wena, okuhlanganisa ukukhiqiza izivivinyo nokuhlola izixazululo zabo, ngaphandle komqeqeshi ongumuntu.

• Abasizi be-AI beKlasi: Emakilasini angokoqobo noma asebenza nge-inthanethi, i-AI ingalalela izingxoxo zekilasi futhi isize uthisha ngokushesha (isb., iziphakamiso zokuhleba nge-earpiece: “Abafundi abaningana babukeka bedidekile ngalowo mqondo, mhlawumbe banikeze esinye isibonelo”). Ingaphinde ilawule izinkundla zekilasi eziku-inthanethi, iphendule imibuzo eqondile ebuzwa ngabafundi (“Isabelo sifika nini?” noma icacise iphuzu lesifundo) ukuze uthisha angahlaselwa ngama-imeyili. Ngo-2035, ukuba nomfundisi osebenza naye we-AI egumbini, kuyilapho uthisha ongumuntu egxila esiqondisweni sezinga eliphezulu kanye nezici zokugqugquzela, kungaba yinto ejwayelekile.

• Ukufinyelela Emfundweni Yomhlaba Wonke: Abafundisi be-AI abazimele bangasiza ekufundiseni abafundi ezindaweni lapho othisha beshoda khona. Ithebhulethi enomfundisi we-AI ingase isebenze njengomfundisi oyinhloko wabafundi abanemfundo elinganiselwe, ihlanganisa ukufunda nokubhala okuyisisekelo kanye nezibalo. Ngo-2035, lokhu kungaba okunye kokusetshenziswa okunomthelela omkhulu - i-AI yokuvala izikhala lapho othisha abangabantu bengatholakali khona. Kodwa-ke, ukuqinisekisa ikhwalithi kanye nokufaneleka kwamasiko kwemfundo ye-AI ezimweni ezahlukene kuzobaluleka.

Ingabe i-AI izothatha indawo yothisha? Cishe ngokuphelele. Ukufundisa kungaphezu kokuletha okuqukethwe - ukuqeqesha, ugqozi, ukwesekwa kwezenhlalo-ngokomzwelo. Lezo zinto zabantu kunzima ukuthi i-AI izilingise. Kodwa i-AI ingaba uthisha wesibili ekilasini noma ngisho nothisha wokuqala wokudluliselwa kolwazi, okushiya othisha abangabantu begxile kulokho abantu abakwenzayo kangcono: ukuzwelana, ukugqugquzela, nokukhuthaza ukucabanga okubucayi.

Kunezinto okufanele ziphathwe: ukuqinisekisa ukuthi i-AI inikeza ulwazi olunembile (akukho ukuqagela kwezemfundo ngamaqiniso angamanga), ukugwema ukubandlulula kokuqukethwe kwezemfundo, ukugcina ubumfihlo bedatha yabafundi, nokugcina abafundi behilelekile (i-AI idinga ukuba yisikhuthazo, hhayi nje ukuthi ilungile). Cishe sizobona ukugunyazwa noma ukuqinisekiswa kwezinhlelo zemfundo ze-AI - okufana nezincwadi zezifundo kuvunyelwa - ukuqinisekisa ukuthi ziyahlangabezana nezindinganiso.

Enye inselele ukuthembela ngokweqile: uma umfundisi we-AI enikeza izimpendulo kalula kakhulu, abafundi bangase bangafundi ukubekezela noma ukuxazulula izinkinga. Ukuze kuncishiswe lokhu, abafundisi be-AI besikhathi esizayo bangase baklanywe ukuze ngezinye izikhathi bavumele abafundi balwe (njengoba umfundisi ongumuntu engenza) noma babakhuthaze ukuba baxazulule izinkinga ngamacebiso kunokuba banikeze izixazululo.

Ngo-2035, ikilasi lingase lishintshe: umfundi ngamunye onedivayisi exhunywe ku-AI ebaqondisa ngejubane lakhe, kuyilapho uthisha ehlela imisebenzi yeqembu futhi enikeza ukuqonda komuntu. Imfundo ingaba ngcono futhi ifaneleke. Isithembiso ukuthi wonke umfundi uthola usizo aludingayo lapho eludinga - ulwazi "lomfundisi womuntu siqu" lwangempela ngobukhulu. Ingozi ukulahlekelwa ukuthintwa komuntu noma ukusebenzisa kabi i-AI (njengabafundi abakopela nge-AI). Kodwa-ke, uma iphathwa kahle, i-AI ekhiqizayo izokwazi ukwenza intando yeningi futhi ithuthukise ukufunda ngokuba ngumngane ohlala ekhona nonolwazi ohambweni lomfundi lwemfundo.

I-AI Ekhiqizayo ku-Logistics kanye ne-Supply Chain

I-Logistics – ubuciko nesayensi yokuthutha izimpahla nokuphatha izintambo zokuhlinzeka – kungase kungabonakali njengesizinda sendabuko se-AI “ekhiqizayo”, kodwa ukuxazulula izinkinga nokuklama okubalulekile kubalulekile kulo mkhakha. I-AI ekhiqizayo ingasiza ngokulingisa izimo, ukwenza ngcono izinhlelo, ngisho nokulawula izinhlelo zamarobhothi. Umgomo we-logistics ukusebenza kahle nokonga izindleko, okuhambisana kahle namandla e-AI ekuhlaziyeni idatha nasekuphakamiseni izixazululo. Ngakho-ke i-AI ingangena kanjani ngokuzimela ekusebenzeni izintambo zokuhlinzeka kanye nemisebenzi ye-logistics?

Amakhono Amanje (2025): Ukuthuthukisa Nokuhlela Ngokuqondiswa Kwabantu

Namuhla, i-AI (kufaka phakathi ezinye izindlela zokukhiqiza) isetshenziswa kakhulu kwezokuthutha njengethuluzi lokusekela izinqumo :

• Ukuthuthukisa Imizila: Izinkampani ezifana ne-UPS ne-FedEx sezivele zisebenzisa ama-algorithms e-AI ukuze zithuthukise imizila yokulethwa - ukuqinisekisa ukuthi abashayeli bathatha indlela esebenza kahle kakhulu. Ngokwesiko lawa kwakuyi-algorithms yocwaningo lwemisebenzi, kodwa manje izindlela zokukhiqiza zingasiza ekuhloleni amasu okuhambisa amanye ngaphansi kwezimo ezahlukahlukene (ithrafikhi, isimo sezulu). Ngenkathi i-AI isikisela imizila, abathumeli noma abaphathi abangabantu babeka amapharamitha (isb., izinto eziza kuqala) futhi bangadlula uma kudingeka.

• Ukuhlela Umthwalo Nesikhala: Kumaloli okupakisha noma izitsha zokuthumela, i-AI ingakhiqiza izinhlelo zokulayisha ezifanele (ibhokisi eliya kuphi). I-AI ekhiqizayo ingakhiqiza ukucushwa okuningi kokupakisha ukuze kwandiswe ukusetshenziswa kwesikhala, empeleni "kudale" izixazululo abantu abangakhetha kuzo. Lokhu kwaqokonyiswa ucwaningo oluphawula ukuthi amaloli avame ukuba nama-30% angenalutho e-US, futhi ukuhlela okungcono - ngosizo lwe-AI - kunganciphisa lokho kungcola ( Ama-Top Generative AI Use Cases in Logistics ). Lezi zinhlelo zokulayisha ezikhiqizwa yi-AI zihlose ukunciphisa izindleko zikaphethiloli kanye nokukhishwa kwegesi, futhi kwezinye izindawo zokugcina impahla zenziwa ngezinguquko ezincane ezenziwe ngesandla.

• Ukubikezela Isidingo Nokuphathwa Kwempahla: Amamodeli e-AI angabikezela isidingo somkhiqizo futhi akhiqize izinhlelo zokubuyisela impahla. Imodeli yokukhiqiza ingase ilingise izimo ezahlukene zesidingo (ake sithi, i-AI "icabanga" ukwanda kwesidingo ngenxa yeholide elizayo) futhi ihlele impahla ngokufanele. Lokhu kusiza abaphathi be-supply chain ukulungiselela. Njengamanje, i-AI inikeza izibikezelo neziphakamiso, kodwa abantu bavame ukwenza isinqumo sokugcina mayelana namazinga okukhiqiza noma uku-oda.

• Ukuhlolwa Kwengozi: Uchungechunge lokuphakelwa komhlaba wonke lubhekene nokuphazamiseka (izinhlekelele zemvelo, ukubambezeleka kwamachweba, izinkinga zezepolitiki). Izinhlelo ze-AI manje zihlola izindaba nedatha ukuze zithole izingozi eziseduze. Isibonelo, enye inkampani yezokuthutha isebenzisa i-gen AI ukuskena i-inthanethi nokumaka imizila yezokuthutha eyingozi (izindawo ezingase zibe nenkinga ngenxa, ake sithi, isiphepho noma isiyaluyalu esizayo) ( Izimo Eziphezulu Zokusetshenziswa Kwe-AI Ku-Logistics ). Ngalolo lwazi, abahleli bangakwazi ukushintsha ngokuzimela ukuthunyelwa ezindaweni ezinenkinga. Kwezinye izimo, i-AI ingase income ngokuzenzakalelayo izinguquko zomzila noma indlela yokushintsha ezokuthutha, abantu abazoyivuma.

• Ukuzenzakalela Kwendawo Yokugcina Izinto: Izindawo eziningi zokugcina izinto zisebenza ngokuzenzakalelayo ngamarobhothi okuqoqa nokupakisha. I-AI ekhiqizayo ingabela imisebenzi kumarobhothi nakubantu ngokuguquguqukayo ukuze kugeleze kahle. Isibonelo, i-AI ingase idale ulayini wemisebenzi wabaqoqi bamarobhothi njalo ekuseni ngokusekelwe kuma-oda. Lokhu kuvame ukuzimela ngokuphelele ekusebenzeni, lapho abaphathi beqapha ama-KPI kuphela - uma ama-oda enyuka ngokungalindelekile, i-AI ilungisa ukusebenza kwayo yodwa.

• Ukuphathwa Kwezimoto: I-AI iyasiza ekuhleleni ukulungiswa kwezimoto ngokuhlaziya amaphethini nokukhiqiza amashejuli okulungiswa afanele anciphisa isikhathi sokungasebenzi. Ingaphinde ihlanganise ukuthunyelwa ukuze kuncishiswe uhambo. Lezi zinqumo zingenziwa yisofthiwe ye-AI ngokuzenzakalelayo uma nje ihlangabezana nezidingo zesevisi.

Sekukonke, kusukela ngo-2025, abantu babebeka imigomo (isib., “ukunciphisa izindleko kodwa baqinisekise ukulethwa kwezinsuku ezi-2”) futhi i-AI ikhipha izixazululo noma amashejuli ukuze kufezwe lokho. Izinhlelo zingasebenza usuku nosuku ngaphandle kokungenelela kuze kube yilapho kwenzeka into engavamile. Izinto eziningi eziphathelene nokuthutha zihilela izinqumo eziphindaphindwayo (lokhu kuthunyelwa kufanele kusuke nini? Yiliphi i-warehouse ukuze kufezwe lo myalelo?), i-AI engafunda ukuyenza njalo. Izinkampani kancane kancane zithemba i-AI ukuthi izosingatha lezi zinqumo ezincane futhi zixwayisa abaphathi kuphela lapho kuvela okuhlukile.

Ithemba lika-2030-2035: Amaketanga Okuhlinzeka Ngomuntu Ozishayelayo

Eminyakeni eyishumi ezayo, singabona ukuhlanganiswa okuzimele kwezokuthutha okuqhutshwa yi-AI:

• Izimoto Ezizimele Nama-Drone: Amaloli azishayela wona kanye nama-drone okulethwa, kuyilapho isihloko esibanzi se-AI/robotics, sithinta ngqo ezokuthutha. Ngo-2030, uma izinselele zomthetho nezobuchwepheshe zinqotshwa, singase sibe nama-AI ashayela amaloli emigwaqweni emikhulu njalo noma ama-drone aphatha ukulethwa kwebanga lokugcina emadolobheni. Lawa ma-AI azokwenza izinqumo zesikhathi sangempela (ukushintsha komzila, ukugwema izithiyo) ngaphandle kwabashayeli abangabantu. I-engeli yokukhiqiza iwukuthi lawa ma-AI ezimoto afunda kanjani kudatha enkulu kanye nokulingisa, ngempumelelo "ukuqeqeshwa" ezimweni ezingenakubalwa. Imikhumbi ezimele ngokugcwele ingasebenza amahora angama-24 ngosuku, izinsuku eziyi-7 ngesonto, abantu beqapha kude kuphela. Lokhu kususa isici esikhulu sabantu (abashayeli) ekusebenzeni kwezokuthutha, okwandisa kakhulu ukuzimela.

• Amaketanga Okuhlinzeka Ngokuzimela: I-AI ekhiqizayo cishe izosetshenziselwa ukulingisa izimo zeketanga lokuhlinzeka njalo nokulungiselela izinhlelo zesimo esiphuthumayo. Ngo-2035, i-AI ingase ibone ngokuzenzakalelayo lapho ifektri yomhlinzeki ivaliwe (ngezindaba noma ama-data feeds) bese ngokushesha ekutholeni abahlinzeki abahlukile esevele ihlolisisiwe ekulingiseni. Lokhu kusho ukuthi iketanga lokuhlinzeka "liyazelapha" ekuphazamisekeni nge-AI ethatha isinyathelo. Abaphathi abangabantu bazokwaziswa ngalokho okwenziwa yi-AI, kunokuba kube yilabo abaqala isixazululo.

• Ukuthuthukiswa Kwempahla Ephelele: I-AI ingaphatha impahla ngokuzimela kuyo yonke inethiwekhi yezindawo zokugcina impahla nezitolo. Inganquma ukuthi izothutha nini futhi kuphi impahla (mhlawumbe isebenzisa amarobhothi noma izimoto ezizenzakalelayo ukwenza kanjalo), igcine impahla eyanele endaweni ngayinye. I-AI ngokuyisisekelo iqhuba umbhoshongo wokulawula uchungechunge lokuhlinzeka: ukubona konke ukugeleza nokwenza izinguquko ngesikhathi sangempela. Ngo-2035, umqondo wochungechunge lokuhlinzeka "oluzishayelayo" ungasho ukuthi uhlelo luzobona uhlelo lokusabalalisa olungcono kakhulu nsuku zonke, lu-ode imikhiqizo, luhlele ukusebenza kwefektri, futhi luhlele ukuthutha konke ngokwalo. Abantu bazosingatha isu eliphelele futhi babhekane nezimo ezingaphandle kokuqonda kwamanje kwe-AI.

• Umklamo Okhiqizayo Kwezokuthutha: Singabona i-AI iklama amanethiwekhi amasha ochungechunge lokuhlinzeka. Ake sithi inkampani inwebeka iye esifundeni esisha; i-AI ingakhiqiza izindawo zokugcina impahla ezifanele, izixhumanisi zokuhamba, kanye nezinqubomgomo zempahla zaleso sifunda uma sinikezwe idatha - into abeluleki nabahlaziyi abayenzayo namuhla. Ngo-2030, izinkampani zingase zithembele ezincomweni ze-AI zokukhetha uchungechunge lokuhlinzeka, zithembele ukuthi izokala izinto ngokushesha futhi mhlawumbe ithole izixazululo zokudala (njengezindawo zokusabalalisa ezingabonakali) abantu abangaziboni.

• Ukuhlanganiswa Nokukhiqiza (Imboni 4.0): Ukuthutha akuzimele wodwa; kuhlangana nokukhiqiza. Amafektri esikhathi esizayo angase abe nokuhlela ukukhiqizwa kwe-AI ekhiqizayo, a-ode izinto zokusetshenziswa ngesikhathi esifanele, bese eyala inethiwekhi yezokuthutha ukuthi ithumele imikhiqizo ngokushesha. Le-AI ehlanganisiwe ingasho ukuhlela okuncane kwabantu iyonke - uchungechunge olungenamthungo kusukela ekukhiqizeni kuya ekulethweni oluqhutshwa yi-algorithms ethuthukisa izindleko, isivinini, kanye nokusimama. Kakade, ngo-2025, izintambo zokuhlinzeka ezisebenza kahle kakhulu seziqhutshwa idatha; ngo-2035 kungenzeka ukuthi seziqhutshwa kakhulu yi-AI.

• Isevisi Yamakhasimende Eguquguqukayo Kwezokuthutha: Ngokusekelwe ku-AI yesevisi yamakhasimende, ama-AI ochungechunge lokuhlinzeka angase axhumane ngqo namakhasimende noma amaklayenti. Isibonelo, uma iklayenti elikhulu lifuna ukushintsha i-oda lalo elikhulu ngomzuzu wokugcina, i-ejenti ye-AI ingaxoxisana ngezindlela ezingenzeka (njengokuthi “Singaletha ingxenye manje, ingxenye ngesonto elizayo ngenxa yemingcele”) ngaphandle kokulinda umphathi ongumuntu. Lokhu kuhilela ukuqonda i-AI ekhiqizayo zombili izinhlangothi (isidingo samakhasimende vs. amandla okusebenza) nokwenza izinqumo ezigcina imisebenzi ibushelelezi ngenkathi yanelisa amaklayenti.

Inzuzo elindelekile olusebenza kahle, oluqinile, nolusabelayo . Izinkampani zibikezela ukonga okukhulu - uMcKinsey ulinganisele ukuthi ukulungiswa kweketanga lokuhlinzeka okuqhutshwa yi-AI kunganciphisa kakhulu izindleko futhi kuthuthukise amazinga esevisi, kungeze inani elingaba yizigidigidi kuzo zonke izimboni ( Isimo se-AI ngo-2023: Unyaka wokuqala we-AI okhiqizayo | uMcKinsey ).

Kodwa-ke, ukuguqula ukulawula okwengeziwe ku-AI nakho kunezingozi, njengokuhluleka ukuqeda amaphutha uma ukucabanga kwe-AI kunephutha (isb., isimo esibi kakhulu se-AI supply chain eqhuba inkampani ngephutha ngenxa yephutha lokumodela). Izivikelo ezifana "nomuntu ongaphakathi kwezinqumo ezinkulu" noma okungenani amadeshibhodi avumela ukuguqulwa okusheshayo komuntu cishe zizohlala kuze kube ngu-2035. Ngokuhamba kwesikhathi, njengoba izinqumo ze-AI zifakazela, abantu bazokhululeka kakhulu ukubuyela emuva.

Ngokuthakazelisayo, ngokwenza ngcono ukuze kusebenze kahle, i-AI ngezinye izikhathi ingenza izinqumo ezingqubuzana nezintandokazi zabantu noma imikhuba yendabuko. Isibonelo, ukwenza ngcono kuphela kungaholela ezinqolobaneni ezithambile kakhulu, okusebenzayo kodwa okungase kubonakale kuyingozi. Ochwepheshe be-supply chain ngo-2030 kungadingeka balungise ukuqonda kwabo ngoba i-AI, eqoqa idatha enkulu, ingabonisa ukuthi isu layo elingavamile lisebenza kangcono.

Okokugcina, kumele sicabangele ukuthi imikhawulo engokwenyama (ingqalasizinda, isivinini senqubo engokwenyama) inciphisa ukuthi izinto ezisetshenziswayo zingashintsha ngokushesha kangakanani, ngakho-ke inguquko lapha imayelana nokuhlela okuhlakaniphile nokusetshenziswa kwezimpahla kunokuba kube ngokoqobo okusha okungokoqobo. Kodwa ngisho nangaphakathi kwaleyo mingcele, izixazululo zokudala ze-AI yokukhiqiza kanye nokwenza ngcono okungapheli kungathuthukisa kakhulu indlela izimpahla ezihamba ngayo emhlabeni wonke ngokuhlela okuncane ngesandla.

Ngamafuphi, ezokuthutha ngo-2035 zingase zisebenze njengomshini ozenzakalelayo onamafutha amaningi: izimpahla zihamba kahle, imizila ishintsha ngesikhathi sangempela ukuze iphazamiseke, izindawo zokugcina impahla ziziphethe ngamarobhothi, kanye nohlelo lonke olufunda futhi luthuthuka njalo kusuka kudatha - konke kuhlelwe yi-AI ekhiqizayo esebenza njengobuchopho bomsebenzi.

I-AI Ekhiqizayo Kwezezimali Namabhizinisi

Imboni yezezimali igxile kakhulu kulwazi - imibiko, ukuhlaziywa, ukuxhumana namakhasimende - okwenza kube yindawo evundile ye-AI ekhiqizayo. Kusukela ebhange kuya ekuphathweni kokutshalwa kwezimali kanye nomshuwalense, izinhlangano zihlola i-AI ukuze zenze izinto ngokuzenzakalelayo kanye nokukhiqiza ulwazi. Umbuzo uwukuthi, yimiphi imisebenzi yezezimali i-AI engasingathwa ngokwethembeka ngaphandle kokubhekwa ngabantu, uma kubhekwa ukubaluleka kokunemba nokwethenjwa kulesi sizinda?

Amakhono Amanje (2025): Imibiko Ezenzakalelayo Nokusekelwa Kwezinqumo

Njenganamuhla, i-AI ekhiqizayo inikela ngezimali ngezindlela eziningana, ngokuvamile ngaphansi kokuqondisa komuntu:

• Ukukhiqizwa Kwemibiko: Amabhange nezinkampani zezimali zikhiqiza imibiko eminingi - izifinyezo zemali engenayo, ukuphawula kwemakethe, ukuhlaziywa kwephothifoliyo, njll. I-AI isivele isetshenziswa ukubhala lokhu. Isibonelo, iBloomberg ithuthukise i-BloombergGPT , imodeli enkulu yolimi eqeqeshwe ngedatha yezezimali, ukusiza ngemisebenzi efana nokuhlukaniswa kwezindaba kanye ne-Q&A kubasebenzisi bayo be-terminal ( i-AI Ekhiqizayo iyeza ukuzoxhasa ngezimali ). Ngenkathi ukusetshenziswa kwayo okuyinhloko ukusiza abantu ukuthola ulwazi, kukhombisa indima ekhulayo ye-AI. I-Automated Insights (inkampani i-AP esebenza nayo) nayo ikhiqize izihloko zezimali. Izincwadi eziningi zezindaba zokutshalwa kwezimali zisebenzisa i-AI ukufingqa izinyathelo zemakethe zansuku zonke noma izinkomba zomnotho. Ngokuvamile, abantu babuyekeza lokhu ngaphambi kokuthumela kumakhasimende, kodwa kuyindlela esheshayo yokubhala kunokuba babhale kusukela ekuqaleni.

• Ukuxhumana Kwamakhasimende: Ebhange lokuthengisa, ama-chatbot e-AI aphatha imibuzo yamakhasimende mayelana nezilinganiso ze-akhawunti, ukuthengiselana, noma ulwazi lomkhiqizo (okuhlanganisa nesizinda sesevisi yamakhasimende). Futhi, i-AI ingakhiqiza izincwadi zeseluleko sezezimali noma izithiyo. Isibonelo, i-AI ingase ibone ukuthi ikhasimende lingonga imali bese libhala ngokuzenzakalelayo umlayezo ophakamisa ukuthi lishintshele kolunye uhlobo lwe-akhawunti, oluzophuma ngaphandle kokungenelela komuntu okuncane. Lolu hlobo lokuxhumana okwenziwe ngokwezifiso ngezinga luwukusetshenziswa kwamanje kwe-AI kwezezimali.

• Ukutholwa Kokukhwabanisa Nezexwayiso: I-AI Ekhiqizayo ingasiza ekudaleni izindaba noma izincazelo zeziphambeko ezitholwe izinhlelo zokukhwabanisa. Isibonelo, uma umsebenzi osolisayo uhlatshwe umkhosi, i-AI ingase ikhiqize umlayezo wencazelo kumakhasimende (“Siqaphele ukungena ngemvume kusuka kudivayisi entsha…”) noma umbiko wabahlaziyi. Ukutholwa kwenziwa ngokuzenzakalela (kusetshenziswa ukutholwa kwe-AI/ML okungavamile), futhi ukuxhumana kuya ngokuya kuzenzakalelayo, yize izenzo zokugcina (ukuvimba i-akhawunti) zivame ukuba nokuhlolwa komuntu.

• Ukwelulekwa Ngezezimali (kunqunyelwe): Abanye abeluleki be-robo (amapulatifomu okutshalwa kwezimali okuzenzakalelayo) basebenzisa ama-algorithms (hhayi ngempela i-AI yokukhiqiza) ukuphatha amaphothifoliyo ngaphandle kwabeluleki babantu. I-AI yokukhiqiza ingena, ake sithi, ngokukhiqiza ukuphawula ngokuthi kungani kwenziwe ukuhweba okuthile noma isifinyezo sokusebenza kwephothifoliyo okwenzelwe iklayenti. Kodwa-ke, iseluleko sezezimali esimsulwa (njengokuhlela kwezezimali okuyinkimbinkimbi) sisalokhu si-algorithmic yabantu noma esekelwe emithethweni; iseluleko sokukhiqiza esikhululekile ngaphandle kokubhekwa siyingozi ngenxa yesibopho uma singalungile.

• Ukuhlolwa Kwengozi Nokubhala Ngobungcweti: Izinkampani zomshuwalense zihlola i-AI ukuze zibhale ngokuzenzakalelayo imibiko yokuhlola ingozi noma ngisho nemibhalo yenqubomgomo ebhaliwe. Isibonelo, uma unikezwe idatha mayelana nempahla, i-AI ingakhiqiza inqubomgomo yomshuwalense ebhaliwe noma umbiko womqinisekisi ochaza izici zengozi. Abantu okwamanje babuyekeza le miphumela ngoba noma yiliphi iphutha enkontilekeni lingabiza kakhulu.

• Ukuhlaziywa Kwedatha Nokuqonda: I-AI ingahlola izitatimende zezimali noma izindaba futhi ikhiqize izifinyezo. Abahlaziyi basebenzisa amathuluzi angafingqa ngokushesha umbiko wonyaka onamakhasi ayi-100 abe amaphuzu abalulekile, noma akhiphe okubalulekile okuvela kumbhalo wezingcingo zenzuzo. Lezi zifinyezo zisindisa isikhathi futhi zingasetshenziswa ngqo ekwenzeni izinqumo noma zidluliselwe phambili, kodwa abahlaziyi abahlakaniphile bahlola kabili imininingwane ebalulekile.

Empeleni, i-AI yamanje kwezezimali isebenza njengomhlaziyi/umbhali ongakhathali , ikhiqiza okuqukethwe okucolisiswa ngabantu. Ukusetshenziswa ngokuzimela ngokugcwele kwenzeka kakhulu ezindaweni ezichazwe kahle njengezindaba eziqhutshwa idatha (akudingeki ukwahlulela okuzimele) noma izimpendulo zesevisi yamakhasimende. Ukuthembela ngqo ku-AI ngezinqumo mayelana nemali (njengokuhambisa izimali, ukwenza ukuhweba ngale kwama-algorithms asethwe ngaphambilini) akuvamile ngenxa yezibalo eziphezulu kanye nokuhlolwa kwemithetho.

Ithemba lika-2030-2035: Abahlaziyi be-AI kanye Nemisebenzi Yezezimali Ezimele

Uma sibheka phambili, ngo-2035 i-AI ekhiqizayo ingase igxile kakhulu ekusebenzeni kwezezimali, okungenzeka iphathe imisebenzi eminingi ngokuzimela:

• Abahlaziyi Bezezimali be-AI: Singase sibone izinhlelo ze-AI ezingahlaziya izinkampani nezimakethe futhi zikhiqize izincomo noma imibiko ezingeni lomhlaziyi wocwaningo lwe-human equity. Ngo-2030, i-AI ingafunda wonke amafayela ezezimali enkampani, iqhathanise nedatha yemboni, futhi ikhiqize umbiko wokuncoma wokutshalwa kwezimali ("Thenga/Thengisa" ngokucabanga) yodwa. Ezinye izikhwama ze-hedge sezivele zisebenzisa i-AI ukukhiqiza izimpawu zokuhweba; ngawo-2030, imibiko yocwaningo lwe-AI ingaba yinto evamile. Abaphathi bephothifoliyo yabantu bangase baqale ukwethemba ukuhlaziywa okukhiqizwe yi-AI njengokufaka okukodwa phakathi kokunye. Kukhona ngisho nethuba lokuthi i-AI iphathe amaphothifoliyo ngokuzimela: ukuqapha njalo nokulinganisa kabusha ukutshalwa kwezimali ngokwesu elichazwe kusengaphambili. Eqinisweni, ukuhweba nge-algorithmic sekuvele kuzenzakalelayo kakhulu - i-AI ekhiqizayo ingenza amasu avumelane kakhulu ngokukhiqiza nokuhlola amamodeli amasha okuhweba ngokwawo.

• Ukuhlela Ezezimali Okuzenzakalelayo: Abeluleki be-AI ababhekene nabathengi bangaphatha ukuhlela kwezezimali okuvamile kwabantu ngabanye. Ngo-2030, ungase utshele i-AI imigomo yakho (ukuthenga indlu, ukonga imali yekolishi) futhi ingakhiqiza uhlelo oluphelele lwezezimali (isabelomali, ukwabiwa kokutshalwa kwezimali, iziphakamiso zomshuwalense) oluqondiswe kuwe. Ekuqaleni umhleli wezezimali ongumuntu angase alubukeze, kodwa njengoba ukuzethemba kukhula, iseluleko esinjalo singanikezwa ngqo kubathengi, ngezithibelo ezifanele. Isihluthulelo kuzoba ukuqinisekisa ukuthi iseluleko se-AI sihambisana nemithetho futhi sizuzisa iklayenti. Uma sixazululiwe, i-AI ingenza iseluleko sezezimali esiyisisekelo sitholakale kalula ngezindleko eziphansi.

• Ukuzenzakalela Kwasehhovisi Lasemuva: I-AI Ekhiqizayo ingase iphathe ngokuzimela imibhalo eminingi yasehhovisi lasemuva - izicelo zemalimboleko, imibiko yokuthobela imithetho, izifinyezo zokuhlolwa. Isibonelo, i-AI ingathatha yonke idatha yokuthengiselana futhi ikhiqize umbiko wokuhlolwa oveza noma yiziphi izinkinga. Abahloli bamabhuku ngo-2035 bangase bachithe isikhathi esiningi bebuyekeza okuhlukile okuphawulwe yi-AI kunokuhlola konke ngokwabo. Ngokufanayo, ukuze kuhambisane nemithetho, i-AI ingakhiqiza imibiko yemisebenzi esolisayo (ama-SAR) yabalawuli ngaphandle kokuba umhlaziyi ayibhale kusukela ekuqaleni. Ukukhiqizwa okuzenzakalelayo kwale mibhalo evamile, kanye nokubhekwa kwabantu okuya esisekelweni sokuhlukile, kungaba yindinganiso.

• Izimangalo Zomshwalense Nokubhala Ngomshuwalense: I-AI ingacubungula isimangalo somshwalense (ngobufakazi besithombe, njll.), inqume ukumbozwa, futhi ikhiqize incwadi yesinqumo sokukhokha ngokuzenzakalelayo. Singase sifinyelele eqophelweni lapho izimangalo eziqondile (njengezingozi zezimoto ezinedatha ecacile) zixazululwa ngokuphelele yi-AI zingakapheli imizuzu yokuthunyelwa. Izinqubomgomo ezintsha zokubhala ngomshuwalense zingafana: I-AI ihlola ubungozi futhi ikhiqize imigomo yenqubomgomo. Ngo-2035, mhlawumbe amacala ayinkimbinkimbi noma angenamkhawulo kuphela adluliselwa kubabhali bomshuwalense abangabantu.

• Ukukhwabanisa Nokuphepha: I-AI cishe izoba yinto ebaluleke kakhulu ekutholeni nasekuphenduleni ukukhwabanisa noma izinsongo ze-cyber kwezezimali. Ama-ejenti e-AI azimele angase aqaphe ukuthengiselana ngesikhathi sangempela futhi athathe izinyathelo ezisheshayo (ukuvimba ama-akhawunti, ukumisa ukuthengiselana) lapho kufinyelela imigomo ethile, bese eveza isizathu. Isivinini sibalulekile lapha, ngakho-ke kudingeka ukubandakanyeka okuncane kwabantu. Ingxenye yokukhiqiza ingase ibe sekuxhumaneni namakhasimende noma nabalawuli ngalezi zenzo ngendlela ecacile.

• Ukusekelwa Kwabaphathi: Cabanga “ngomphathi wabasebenzi” be-AI ongakhiqiza imibiko yebhizinisi yabaphathi ngokushesha. Buza, “Isigaba sethu saseYurophu senze kanjani kule kota futhi yiziphi izinto ezibangela inkinga uma kuqhathaniswa nonyaka odlule?” bese i-AI izokhiqiza umbiko omfushane onamashadi, wonke enembile, edonsa idatha. Lolu hlobo lokubika okuguquguqukayo, okuzimele kanye nokuhlaziya kungaba lula njengengxoxo. Ngo-2030, ukubuza ubuhlakani bebhizinisi be-AI nokuthembela ukuthi bunikeze izimpendulo ezifanele kungathatha indawo yemibiko engaguquki futhi mhlawumbe ngisho nezinye izindima zomhlaziyi.

Ukuqagela okukodwa okuthakazelisayo: ngawo-2030, iningi lokuqukethwe kwezezimali (izindaba, imibiko, njll.) kungase kube okukhiqizwa yi-AI . Kakade, izitolo ezifana ne-Dow Jones ne-Reuters zisebenzisa ukuzenzakalela kwezinye izindaba. Uma lowo mkhuba uqhubeka, futhi uma kubhekwa ukwanda kwedatha yezezimali, i-AI ingaba nesibopho sokuhlunga nokudlulisa iningi layo.

Kodwa-ke, ukwethembana nokuqinisekisa kuzoba yinto ebalulekile. Imboni yezezimali ilawulwa kakhulu futhi noma iyiphi i-AI esebenza ngokuzimela kuzodingeka ihlangabezane nezindinganiso eziqinile:

• Ukuqinisekisa ukuthi akukho ukubona izinto ezingekho (awukwazi ukuthola umhlaziyi we-AI ukuthi asungule isilinganiso sezezimali esingelona iqiniso - esingadukisa izimakethe).

• Ukugwema ubandlululo noma imikhuba engekho emthethweni (njengokwenza izinqumo zokubolekisa zibe zimbi ngokungahlosile ngenxa yedatha yokuqeqeshwa ebandlululayo).

• Ukuhlolwa: abalawuli cishe bazofuna ukuthi izinqumo ze-AI zichazwe. Uma i-AI yenqaba imali mboleko noma yenza isinqumo sokuhweba, kumele kube nesizathu esingahlolwa. Amamodeli akhiqizayo angaba yibhokisi elimnyama, ngakho-ke lindela ukuthuthukiswa kwamasu e-AI achazekayo ukuze wenze izinqumo zawo zibe sobala.

Iminyaka eyi-10 ezayo cishe izobandakanya ukubambisana okuseduze phakathi kwe-AI kanye nochwepheshe bezezimali, kancane kancane kushukumise umugqa wokuzimela njengoba ukuzethemba kukhula. Ukuphumelela kwasekuqaleni kuzofika ezenzakalelayo ezinobungozi obuphansi (njengokukhiqizwa kwemibiko). Kuzoba nzima kakhulu izahlulelo eziyinhloko njengezinqumo zesikweletu noma ukukhetha ukutshalwa kwezimali, kodwa ngisho nalapho, njengoba irekhodi le-AI lakhiwa, izinkampani zingase ziyinike ukuzimela okwengeziwe. Isibonelo, mhlawumbe isikhwama se-AI sizosebenza nomqondisi ongumuntu ongenelela kuphela uma ukusebenza kuphazamiseka noma uma i-AI iveza ukungaqiniseki.

Ngokomnotho, uMcKinsey ulinganisele ukuthi i-AI (ikakhulukazi i-gen AI) ingangeza inani lamadola ayizigidigidi ezingama-200-340 ebhange minyaka yonke kanye nemiphumela emikhulu efanayo emakethe yomshuwalense kanye neyezimali ( Isimo se-AI ngo-2023: Unyaka we-Generative AI oqalayo | McKinsey ) ( Liyini ikusasa le-Generative AI? | McKinsey ). Lokhu kwenzeka ngokusebenza kahle kanye nemiphumela yezinqumo ezingcono. Ukuze kubanjwe lelo nani, ukuhlaziywa okuningi kwezezimali okuvamile kanye nokuxhumana cishe kuzodluliselwa ezinhlelweni ze-AI.

Ngamafuphi, ngo-2035 i-AI ekhiqizayo ingaba njengebutho labahlaziyi abancane, abeluleki, kanye nababhali abasebenza emkhakheni wezezimali, benza umsebenzi omningi wokubhonga kanye nokuhlaziywa okuyinkimbinkimbi ngokuzimela. Abantu basazobeka imigomo futhi baphathe isu eliphakeme, ubudlelwano namakhasimende, kanye nokuqapha. Izwe lezezimali, ngokuqapha, lizokwandisa ukuzimela kancane kancane - kodwa isiqondiso sisobala ukuthi ukucubungula ulwazi okuningi kanye nezincomo zezinqumo kuzovela ku-AI. Okungcono kakhulu, lokhu kuholela ekusetshenzisweni okusheshayo (imali mboleko esheshayo, iseluleko sesikhathi sonke), izindleko eziphansi, kanye nokungakhethi hlangothi okungenzeka kube ngaphezulu (izinqumo ezisekelwe kumaphethini wedatha). Kodwa ukugcina ukwethembana kuzobaluleka; iphutha elilodwa le-AI elivelele kwezezimali lingabangela umonakalo omkhulu (cabanga ngokuphahlazeka kwe-AI okubangelwe yi-AI noma inzuzo enqatshelwe ngokungafanele ezinkulungwaneni zabantu). Ngakho-ke, izithiyo zokuqapha kanye nokuhlolwa kwabantu kungenzeka kuqhubeke ikakhulukazi ezenzweni ezibhekene nabathengi, ngisho noma izinqubo zasehhovisi zizimele kakhulu.

Izinselele kanye Nokucatshangelwa Kokuziphatha

Kuzo zonke lezi zinkundla, njengoba i-AI ekhiqizayo ithatha imithwalo yemfanelo eyengeziwe yokuzimela, kuvela isethi yezinselelo ezivamile kanye nemibuzo yokuziphatha. Ukuqinisekisa ukuthi i-AI iyi-ejenti ezimele ethembekile futhi enenzuzo akuyona nje umsebenzi wezobuchwepheshe, kodwa umsebenzi womphakathi. Lapha sichaza izinkinga ezibalulekile nokuthi zixazululwa kanjani (noma kuzodingeka kubhekwane nazo):

Ukuthembeka Nokunemba

Inkinga Yokubona Izinto Ezingacacile: Amamodeli e-AI akhiqizayo angakhiqiza imiphumela engalungile noma eyenziwe ngokuphelele ebonakala iqinisekile. Lokhu kuyingozi kakhulu uma kungekho muntu onolwazi lokubamba amaphutha. I-chatbot inganikeza ikhasimende imiyalelo engalungile, noma umbiko obhaliwe we-AI ungase uqukathe izibalo ezenziwe. Kusukela ngo-2025, ukungalungi kubhekwa njengengozi ephezulu ye-AI ekhiqizayo yizinhlangano ( Isimo se-AI ngo-2023: Unyaka wokuqhamuka kwe-AI ekhiqizayo | McKinsey ) ( Isimo se-AI: Ucwaningo Lomhlaba Wonke | McKinsey ). Ukuqhubekela phambili, amasu afana nokuhlola amaqiniso ngokumelene nezizindalwazi, ukuthuthukiswa kokwakhiwa kwamamodeli, kanye nokufunda okuqinisiwe ngempendulo kusetshenziswa ukuze kuncishiswe ukubona izinto ezingacacile. Izinhlelo ze-AI ezizimele cishe zizodinga ukuhlolwa okuqinile kanye nokuqinisekiswa okusemthethweni kwemisebenzi ebalulekile (njengokwenziwa kwekhodi okungaletha amaphutha/amaphutha okuphepha uma kungalungile).

Ukuvumelana: Izinhlelo ze-AI zidinga ukusebenza ngokwethembeka ngokuhamba kwesikhathi nakuzo zonke izimo. Isibonelo, i-AI ingenza kahle emibuzweni ejwayelekile kodwa ifinyelele ezimweni ezibucayi. Ukuqinisekisa ukusebenza okulinganayo kuzodinga idatha yokuqeqeshwa ebanzi ehlanganisa izimo ezahlukahlukene kanye nokuqapha okuqhubekayo. Izinhlangano eziningi zihlela ukuba nezindlela ezihlanganisiwe - i-AI iyasebenza, kodwa amasampula angahleliwe ahlolwa ngabantu - ukuze kulinganiswe amazinga okunemba okuqhubekayo.

Ama-Fail-Safes: Uma i-AI izimele, ukuba iqaphele ukungaqiniseki kwayo kubalulekile. Uhlelo kufanele lwakhiwe ukuze “lwazi uma lungazi.” Isibonelo, uma udokotela we-AI engaqiniseki ngokuxilonga, kufanele abeke uphawu lokubuyekezwa komuntu kunokunikeza ukuqagela okungahleliwe. Ukwakha isilinganiso sokungaqiniseki emiphumeleni ye-AI (kanye nokuba nemingcele yokudluliselwa komuntu ngokuzenzakalela) kuyindawo esebenzayo yentuthuko.

Ukukhetha kanye Nokulunga

I-AI ekhiqizayo ifunda kudatha yomlando engaqukatha ukucwasa (ubuhlanga, ubulili, njll.). I-AI ezimele ingase iqhubekisele phambili noma ikhulise lokho kucwasa:

• Ekuqasheni noma ekwamukelweni, umenzi wezinqumo we-AI angabandlulula ngokungafanele uma idatha yakhe yokuqeqeshwa inobandlululo.

• Enkonzweni yamakhasimende, i-AI ingasabela ngendlela ehlukile kubasebenzisi ngokusekelwe kolimi noma ezinye izici ngaphandle kokuthi ihlolwe ngokucophelela.

• Emasimini okudala, i-AI ingase imele amasiko noma izitayela ezithile uma isethi yokuqeqesha yayingalingani.

Ukubhekana nalokhu kudinga ukuhlelwa kwedatha ngokucophelela, ukuhlolwa kokukhetha, kanye nokulungiswa kwe-algorithm ukuqinisekisa ubulungisa. Ukucaca kubalulekile: izinkampani kuzodingeka zidalule izindlela zokuthatha izinqumo ze-AI, ikakhulukazi uma i-AI ezimele ithinta amathuba noma amalungelo omuntu (njengokuthola imali mboleko noma umsebenzi). Abalawuli sebevele benaka; isib., uMthetho we-AI we-EU (owawusasebenza maphakathi nawo-2020) cishe uzodinga ukuhlolwa kokukhetha kwezinhlelo ze-AI ezinobungozi obukhulu.

Ukuziphendulela kanye Nesibopho Sezomthetho

Uma uhlelo lwe-AI olusebenza ngokuzimela lubangela umonakalo noma lwenza iphutha, ubani onecala? Izinhlaka zomthetho ziyafinyelela esiphethweni:

• Izinkampani ezisebenzisa i-AI cishe zizoba necala, okufana nokuba necala ngezenzo zesisebenzi. Isibonelo, uma i-AI inikeza iseluleko esibi sezezimali esiholela ekulahlekelweni, inkampani kungadingeka ikhokhele iklayenti.

• Kukhona impikiswano mayelana "nobuntu" be-AI noma ukuthi i-AI ethuthukisiwe ingaba necala ngokwengxenye, kodwa lokho sekungokwengqondo manje. Empeleni, ukusolwa kuzobuyela emuva kubathuthukisi noma kubaqhubi.

• Imikhiqizo emisha yomshuwalense ingase ivele ngenxa yokwehluleka kwe-AI. Uma iloli elizishayelayo libanga ingozi, umshuwalense womenzi ungase uyikhokhele, okufana nesibopho somkhiqizo.

• Ukubhala phansi nokubhala phansi izinqumo ze-AI kuzobaluleka ekuhlolweni kwezidumbu. Uma kukhona okungahambi kahle, sidinga ukuhlola umkhondo wesinqumo se-AI ukuze sifunde kuwo futhi sinikeze umthwalo wemfanelo. Abalawuli bangagunyaza ukurekhoda ngezenzo ze-AI ezizimele ngenxa yalesi sizathu.

Ukucaca Nokuchazeka

I-Autonomous AI kufanele ikwazi ukuchaza izizathu zayo ngamagama aqondakalayo kubantu, ikakhulukazi ezizindeni ezibalulekile (ezezimali, ezempilo, uhlelo lobulungiswa). I-Explainable AI iwumkhakha ozama ukuvula ibhokisi elimnyama:

• Uma imali mboleko yenqatshwa yi-AI, imithethonqubo (njengase-US, i-ECOA) ingadinga ukuthi umfakisicelo anikeze isizathu. Ngakho-ke i-AI kumele iveze izici (isb., “isilinganiso esiphezulu sesikweletu nomholo”) njengencazelo.

• Abasebenzisi abasebenzisana ne-AI (njengabafundi abanomfundisi we-AI noma iziguli ezine-app yezempilo ye-AI) bafanelwe ukwazi ukuthi ifika kanjani ekwelulekweni. Kuzanywa ukwenza ukucabanga kwe-AI kulandeleke kalula, kungaba ngokwenza amamodeli abe lula noma ngokuba namamodeli achazayo afanayo.

• Ukucaca kusho nokuthi abasebenzisi kufanele bazi nini ne-AI uma kuqhathaniswa nomuntu. Iziqondiso zokuziphatha (futhi mhlawumbe eminye imithetho) zithambekele ekudingeni ukudalulwa uma ikhasimende likhuluma ne-bot. Lokhu kuvimbela ukukhohlisa futhi kuvumela imvume yomsebenzisi. Ezinye izinkampani manje sezimaka ngokusobala okuqukethwe okubhalwe yi-AI (njengokuthi “Lesi sihloko sakhiwe yi-AI”) ukuze kugcinwe ukwethenjwa.

Ubumfihlo kanye Nokuvikelwa Kwedatha

I-AI ekhiqizayo ivame ukudinga idatha - kufaka phakathi idatha yomuntu siqu engase ibe bucayi - ukuze isebenze noma ifunde. Imisebenzi ezenzakalelayo kumele ihloniphe ubumfihlo:

• I-ejenti yesevisi yamakhasimende ye-AI izofinyelela ulwazi lwe-akhawunti ukusiza ikhasimende; ukuthi idatha kumele ivikelwe futhi isetshenziswe kuphela kulo msebenzi.

• Uma abafundisi be-AI benokufinyelela kumaphrofayili abafundi, kunezindlela zokucatshangelwa ngaphansi kwemithetho efana ne-FERPA (e-US) ukuqinisekisa ubumfihlo bedatha yezemfundo.

• Amamodeli amakhulu angakhumbula ngokungaqondile imininingwane ethile evela kudatha yawo yokuqeqeshwa (isb., ukuphinda ikheli lomuntu elibonwe ngesikhathi sokuqeqeshwa). Amasu afana nobumfihlo obuhlukile kanye nokungaziwa kwedatha ekuqeqeshweni kubalulekile ukuvimbela ukuvuza kolwazi lomuntu siqu emiphumeleni ekhiqizwe.

• Imithethonqubo efana ne-GDPR inikeza abantu amalungelo phezu kwezinqumo ezizenzakalelayo ezibathintayo. Abantu bangacela ukubuyekezwa kwabantu noma izinqumo ukuthi zingabi ngokuzenzakalela kuphela uma zibathinta kakhulu. Ngo-2030, le mithethonqubo ingase iguquke njengoba i-AI isabalala kakhulu, mhlawumbe ilethe amalungelo okuchaza noma ikhethe ukuphuma ekucutshungulweni kwe-AI.

Ukuphepha Nokuxhashazwa

Izinhlelo ze-AI ezizimele zingaba yizinhloso zokugenca noma zingasetshenziswa ukwenza izinto ezinonya:

• Umkhiqizi wokuqukethwe kwe-AI angasetshenziswa kabi ukudala ulwazi olungelona iqiniso ngezinga elikhulu (amavidiyo e-deepfake, izihloko zezindaba ezingamanga), okuyingozi emphakathini. Izimiso zokuziphatha zokukhipha amamodeli anamandla kakhulu okukhiqiza zixoxwa kakhulu (i-OpenAI ekuqaleni yayiqaphile ngamakhono esithombe se-GPT-4, isibonelo). Izixazululo zifaka phakathi ukumaka okuqukethwe okukhiqizwe yi-AI ukuze kusizwe ukuthola ama-fake, kanye nokusebenzisa i-AI ukulwa ne-AI (njengezindlela zokuthola ama-deepfake).

• Uma i-AI ilawula izinqubo zomzimba (ama-drone, izimoto, ukulawulwa kwezimboni), ukuyivikela ekuhlaselweni kwe-cyber kubalulekile. Uhlelo oluzimele olugetshengiwe lungabangela umonakalo emhlabeni wangempela. Lokhu kusho ukubethela okuqinile, ukuvikela amaphutha, kanye nekhono lokushintsha noma ukuvala komuntu uma kukhona okubonakala sengathi kusengozini.

• Kukhona futhi ukukhathazeka ngokuthi i-AI idlulele ngale kwemingcele ehlosiwe (isimo "se-AI esibi"). Nakuba ama-AI amanje engenawo amandla noma inhloso, uma izinhlelo ezizimele zesikhathi esizayo zisebenza kakhulu, kudingeka imingcele eqinile kanye nokuqapha ukuqinisekisa ukuthi awenzi, isibonelo, ukuhweba okungagunyaziwe noma ukwephula imithetho ngenxa yenhloso echazwe kabi.

Ukusetshenziswa Kokuziphatha kanye Nomthelela Kubantu

Okokugcina, izinto ezibaluleke kakhulu okufanele uzicabangele:

• Ukufuduka Emsebenzini: Uma i-AI ingenza imisebenzi ngaphandle kokungenela komuntu, kwenzekani kuleyo misebenzi? Ngokomlando, ubuchwepheshe buzenzakalela eminye imisebenzi kodwa budale eminye. Ushintsho lungaba buhlungu kubasebenzi abanamakhono emisebenzini eba ngokuzenzakalelayo. Umphakathi kuzodingeka uphathe lokhu ngokuvuselela amakhono, imfundo, kanye nokucabanga kabusha ngokusekelwa kwezomnotho (abanye basikisela ukuthi i-AI ingadinga imibono efana nemali engenayo eyisisekelo uma umsebenzi omningi uzenzakalelayo). Kakade, izinhlolovo zibonisa imizwa exubile - ucwaningo olulodwa luthole ukuthi ingxenye yesithathu yabasebenzi ikhathazekile nge-AI ethatha indawo yemisebenzi, kanti abanye bayibona njengokususa umsebenzi onzima.

• Ukuguguleka Kwamakhono Abantu: Uma othisha be-AI befundisa futhi abashayeli be-AI abazenzakalelayo beshayela futhi i-AI ibhala ikhodi, ingabe abantu bazolahlekelwa yilawa makhono? Ukuthembela kakhulu ku-AI kungalimaza ubuchwepheshe; kuyinto okufanele kulungiswe yizinhlelo zemfundo nokuqeqesha, ukuqinisekisa ukuthi abantu basafunda izisekelo noma ngabe i-AI iyasiza.

• Ukwenza Izinqumo Zokuziphatha: I-AI ayinakho ukwahlulela kokuziphatha kwabantu. Kwezokunakekelwa kwempilo noma emthethweni, izinqumo ezisekelwe kudatha kuphela zingangqubuzana nozwelo noma ubulungiswa ezimweni ngazinye. Kungadingeka sihlanganise izinhlaka zokuziphatha zibe yi-AI (indawo yocwaningo lokuziphatha lwe-AI, isib. ukuvumelanisa izinqumo ze-AI nezindinganiso zabantu). Okungenani, ukugcina abantu besesimweni sezinqumo ezilawulwa ngokokuziphatha kuyalulekwa.

• Ukubandakanya: Ukuqinisekisa ukuthi izinzuzo ze-AI zisatshalaliswa kabanzi kuwumgomo wokuziphatha. Uma izinkampani ezinkulu kuphela ezingakwazi ukukhokhela i-AI ethuthukisiwe, amabhizinisi amancane noma izifunda ezimpofu zingase zishiywe ngemuva. Imizamo yomthombo ovulekile kanye nezixazululo ze-AI ezingabizi zingasiza ekufinyeleleni ngentando yeningi. Futhi, izixhumi kufanele zakhiwe ukuze noma ubani akwazi ukusebenzisa amathuluzi e-AI (izilimi ezahlukene, ukufinyeleleka kulabo abakhubazekile, njll.), funa sidale ukuhlukana okusha kwedijithali kokuthi “ubani onomsizi we-AI nokuthi ubani ongenaye.”

Ukunciphisa Ingozi Kwamanje: Ngakolunye uhlangothi oluhle, njengoba izinkampani ziqala ukusebenzisa i-gene AI, kuyanda ukuqwashisa kanye nesenzo ngalezi zinkinga. Ngasekupheleni kuka-2023, cishe ingxenye yezinkampani ezisebenzisa i-AI zazisebenza ngenkuthalo ukunciphisa izingozi ezifana nokungalungi ( Isimo se-AI ngo-2023: Unyaka wokuqhamuka kwe-AI ekhiqizayo | McKinsey ) ( Isimo se-AI: Ucwaningo Lomhlaba Wonke | McKinsey ), futhi lelo nani liyakhuphuka. Izinkampani zobuchwepheshe sezimise amabhodi okuziphatha e-AI; ohulumeni babhala imithetho. Isihluthulelo ukubhaka ukuziphatha ekuthuthukisweni kwe-AI kusukela ekuqaleni (“Izimiso ngokuklama”), kunokuba zisabele kamuva.

Ekuphetheni ngezinselelo: ukunikeza i-AI ukuzimela okwengeziwe kuyinkemba esika nhlangothi zombili. Kungaveza ukusebenza kahle kanye nokusungula izinto ezintsha, kodwa kudinga umthwalo wemfanelo omkhulu. Iminyaka ezayo cishe izobona ingxube yezixazululo zobuchwepheshe (zokuthuthukisa ukuziphatha kwe-AI), izixazululo zenqubo (izinhlaka zenqubomgomo kanye nokuqapha), kanye mhlawumbe nezindinganiso noma izitifiketi ezintsha (izinhlelo ze-AI zingahlolwa futhi ziqinisekiswe njengezinjini noma izinto zikagesi namuhla). Ukubhekana ngempumelelo nalezi zinselelo kuzonquma ukuthi singazihlanganisa kanjani i-AI ezizimele emphakathini ngendlela ethuthukisa inhlalakahle yabantu kanye nokwethembana.

Isiphetho

I-AI ekhiqizayo ishintshe ngokushesha isuka ekuhlolweni okusha yaya kubuchwepheshe obujwayelekile obuguqulayo obuthinta yonke indawo yezimpilo zethu. Leli phepha elimhlophe lihlole ukuthi, ngo-2025, izinhlelo ze-AI sezivele zibhala izihloko, ziklama ihluzo, zibhala amakhodi esofthiwe, zixoxa namakhasimende, zifingqa amanothi ezokwelapha, zifundisa abafundi, zithuthukisa izintambo zokuhlinzeka, futhi zibhala imibiko yezezimali. Okubalulekile, emisebenzini eminingi yale i-AI ingasebenza ngaphandle kokungenelela komuntu , ikakhulukazi emisebenzini echazwe kahle, ephindaphindwayo. Izinkampani nabantu ngabanye sebeqala ukuthemba i-AI ukuthi izokwenza le misebenzi ngokuzimela, ithola izinzuzo ngesivinini nangesilinganiso.

Uma sibheke phambili ku-2035, simi onqenqemeni lwesikhathi lapho i-AI izoba ngumhlanganyeli obanzi kakhulu - ngokuvamile abasebenzi bedijithali abangabonakali abaphatha inqubo ukuze abantu bakwazi ukugxila kokumangalisayo. Silindele ukuthi i-AI ekhiqizayo izoshayela izimoto namaloli ngendlela ethembekile emigwaqweni yethu, iphathe impahla ezindaweni zokugcina impahla ubusuku bonke, iphendule imibuzo yethu njengabasizi bomuntu siqu abanolwazi, inikeze imfundo yomuntu ngamunye kubafundi emhlabeni wonke, futhi isize ngisho nokuthola izindlela ezintsha zokwelapha - konke ngokuqapha okuqondile okuncane kakhulu. Umugqa phakathi kwethuluzi ne-ejenti uzofiphala njengoba i-AI isuka ekulandeleni imiyalelo ngokunganaki iye ekukhiqizeni izixazululo ngendlela ephumelelayo.

Kodwa-ke, uhambo oluya kuleli kusasa le-AI elizimele kumele luqondiswe ngokucophelela. Njengoba sesichazile, isizinda ngasinye sinemikhawulo yaso kanye nemithwalo yemfanelo:

• Ukuhlolwa Kwangempela Kwanamuhla: I-AI ayiyona into engenaphutha. Ihamba phambili ekuqaphelweni kwamaphethini nasekukhiqizweni kokuqukethwe kodwa ayinakho ukuqonda kwangempela kanye nokuqonda okuvamile ngomqondo womuntu. Ngakho-ke, okwamanje, ukuqondiswa komuntu kusalokhu kuyinetha lokuphepha. Ukuqaphela ukuthi i-AI ikulungele ukundiza yodwa (futhi lapho ingekho) kubalulekile. Impumelelo eminingi namuhla ivela kumodeli weqembu le-AI yomuntu , futhi le ndlela ehlanganisiwe izoqhubeka nokuba wusizo lapho ukuzimela okugcwele kungakabi nengqondo.

• Isithembiso Sakusasa: Ngokuthuthuka kwezakhiwo zamamodeli, amasu okuqeqesha, kanye nezindlela zokuqapha, amakhono e-AI azoqhubeka nokukhula. Iminyaka eyishumi ezayo ye-R&D ingaxazulula amaphuzu amaningi abuhlungu akhona (ukunciphisa ukubona izinto ezingekho, ukuthuthukisa ukuhunyushwa, ukuvumelanisa i-AI nezindinganiso zabantu). Uma kunjalo, izinhlelo ze-AI ngo-2035 zingaba ziqinile ngokwanele ukuba ziphathiswe ukuzimela okukhulu kakhulu. Ukuqagela okukuleli phepha - kusukela kothisha be-AI kuya emabhizinisini aziphethe kakhulu - kungaba yiqiniso lethu, noma kudlulwe ngisho nezinto ezintsha okunzima ukuzicabanga namuhla.

• Indima Yomuntu Nokuzivumelanisa Nezimo: Esikhundleni sokushintsha i-AI ngokusobala, sibona izindima zishintsha. Ochwepheshe kuzo zonke izinkambu cishe kuzodingeka babe nekhono lokusebenza ne -AI - ukuyiqondisa, ukuyiqinisekisa, nokugxila ezicini zomsebenzi ezidinga amandla omuntu njengozwela, ukucabanga okunesu, kanye nokuxazulula izinkinga eziyinkimbinkimbi. Imfundo nokuqeqeshwa kwabasebenzi kufanele kugxile ekugcizeleleni la makhono omuntu ahlukile, kanye nokufunda nge-AI kuwo wonke umuntu. Abenzi bezinqubomgomo nabaholi bamabhizinisi kufanele bahlele izinguquko emakethe yezabasebenzi futhi baqinisekise izinhlelo zokusekela labo abathintekayo ngokuzenzakalelayo.

• Ukuziphatha Nokuphatha: Mhlawumbe okubaluleke kakhulu, uhlaka lokusetshenziswa kwe-AI yokuziphatha kanye nokuphatha kumele lusekele lokhu kukhula kobuchwepheshe. Ukwethembana kuyindlela yokwamukelwa - abantu bazovumela i-AI ukuthi ishayele imoto noma isize ekuhlinzeni uma bethemba ukuthi kuphephile. Ukwakha lokho kuthembana kuhilela ukuhlolwa okuqinile, ukucaca, ukubandakanyeka kwabathintekayo (isb., ukubandakanya odokotela ekwakheni i-AI yezokwelapha, othisha kumathuluzi emfundo ye-AI), kanye nomthethonqubo ofanele. Ukubambisana kwamazwe ngamazwe kungadingeka ukuze kusingathwe izinselele ezifana ne-deepfakes noma i-AI empini, kuqinisekiswe izindinganiso zomhlaba wonke zokusetshenziswa okunesibopho.

Ekuphetheni, i-AI ekhiqizayo imelela njengenjini enamandla yentuthuko. Uma isetshenziswa ngokuhlakanipha, ingakhulula abantu ekusebenzeni kanzima, ivule ubuhlakani, inikeze izinsizakalo ngokwezifiso, futhi ibhekane nezikhala (ukuletha ubuchwepheshe lapho ochwepheshe bentuleka khona). Isihluthulelo ukuwusebenzisa ngendlela eyandisa amandla abantu kunokuwabeka eceleni . Esikhathini esifushane, lokho kusho ukugcina abantu besesimweni sokuqondisa i-AI. Esikhathini eside, kusho ukufaka amanani obuntu enkabeni yezinhlelo ze-AI ukuze ngisho nalapho besebenza ngokuzimela, benze ngokwezintshisekelo zethu sonke.

Isizinda	Ukuzimela Okuthembekile Namuhla (2025)	Ukuzimela Okuthembekile Okulindelekile ngo-2035
Ukubhala Nokuqukethwe	- Izindaba ezijwayelekile (ezemidlalo, imali engenayo) ezikhiqizwa ngokuzenzakalela.- Ukubuyekezwa komkhiqizo okufingqiwe yi-AI.- Imidwebo yezindatshana noma ama-imeyili okuhlelwa ngabantu. ( Philana Patterson – Iphrofayela Yomphakathi ye-ONA ) ( I-Amazon ithuthukisa ulwazi lokubuyekezwa kwamakhasimende nge-AI )	- Izindaba eziningi kanye nokuqukethwe kokumaketha kubhalwe ngokuzenzakalelayo ngokunemba kwamaqiniso.- I-AI ikhiqiza izihloko eziphelele kanye nokukhishwa kwabezindaba ngaphandle kokubhekwa kakhulu.- Okuqukethwe okwenziwe ngezifiso kakhulu okukhiqizwa ngokufunwa.
Ubuciko Nokuklama Okubonakalayo	- I-AI ikhiqiza izithombe ezivela ezikhuthazweni (umuntu ukhetha okungcono kakhulu).- Ubuciko bomqondo kanye nokuhlukahluka kokuklama kudalwe ngokuzimela.	- I-AI ikhiqiza izigcawu ezigcwele zevidiyo/amafilimu kanye nezithombe eziyinkimbinkimbi.- Umklamo okhiqizayo wemikhiqizo/izakhiwo uhlangabezana nezimfuneko.- Imidiya eyenzelwe wena (izithombe, ividiyo) edalwe ngokufunwa.
Ukubhala Amakhodi Kwesofthiwe	- I-AI iqedela ikhodi ngokuzenzakalelayo futhi ibhale imisebenzi elula (ibuyekezwe yi-dev).- Ukukhiqizwa kokuhlolwa okuzenzakalelayo kanye neziphakamiso zamaphutha. ( Ukubhala ikhodi ku-Copilot: Idatha ka-2023 Iphakamisa Ingcindezi Ephansi Ekhwalithini Yekhodi (kufaka phakathi ukubikezela kuka-2024) - I-GitClear ) ( I-GitHub Copilot Ihamba Phambili Umbiko Wocwaningo ku-AI Code Assistants -- I-Visual Studio Magazine )	- I-AI isebenzisa zonke izici ezivela kuma-specs ngokwethembeka.- Ukulungisa amaphutha okuzenzakalelayo kanye nokugcinwa kwekhodi yamaphethini aziwayo.- Ukudalwa kohlelo lokusebenza olunekhodi ephansi ngokufakwa okuncane komuntu.
Insizakalo yekhasimende	- Ama-Chatbot aphendula imibuzo evame ukubuzwa, axazulule izinkinga ezilula (amacala ayinkimbinkimbi okudluliselwa).- I-AI iphatha ~70% yemibuzo ejwayelekile kwezinye iziteshi. ( Izibalo zesevisi yamakhasimende ze-AI ezingama-59 zango-2025 ) ( Ngo-2030, ama-69% ezinqumo ngesikhathi sokusebenzisana kwamakhasimende azoba ... )	- I-AI iphatha ukuxhumana okuningi kwamakhasimende kusukela ekuqaleni kuya ekugcineni, kufaka phakathi imibuzo eyinkimbinkimbi.- Ukwenza izinqumo ze-AI zesikhathi sangempela zezibonelelo zesevisi (ukubuyiselwa kwemali, ukuthuthukiswa).- Amanxusa abantu kuphela uma kukhushulwa noma kumacala akhethekile.
Ukunakekela impilo	- I-AI ibhala amanothi ezokwelapha; iphakamisa ukuxilongwa odokotela abakuqinisekisayo.- I-AI ifunda ezinye izikena (i-radiology) ngokunganaki; ihlola amacala alula. ( Imikhiqizo Yezithombe Zezokwelapha ze-AI Ingakhula Ngokuphindwe Kahlanu ngo-2035 )	- I-AI ihlola ngokuthembekile izifo ezivamile futhi ihumushe izithombe eziningi zezokwelapha. - I-AI iqapha iziguli futhi iqalise ukunakekelwa (isb., izikhumbuzo zemithi, izaziso eziphuthumayo). - "Abahlengikazi" be-AI abaku-inthanethi baphatha ukulandelwa okuvamile; odokotela bagxila ekunakekelweni okuyinkimbinkimbi.
Imfundo	- Abafundisi be-AI baphendula imibuzo yabafundi, badale izinkinga zokuzijwayeza (abaqaphi bothisha).- I-AI isiza ekutholeni amamaki (ngokubuyekezwa kothisha). ([I-AI Ekhiqizayo yemfundo ye-K-12	Umbiko Wocwaningo ngu-Applify]( https://www.applify.co/research-report/gen-ai-for-k12#:~:text=AI%20tutors%3A%20Virtual%20AI,individual%20learning%20styles%20and%20paces ))
Ukuthutha	- I-AI ithuthukisa imizila yokulethwa kanye nokupakisha (abantu babeka imigomo).- I-AI ikhomba izingozi zochungechunge lokuhlinzeka futhi iphakamise izindlela zokunciphisa. ( Amacala Okusetshenziswa Kwe-AI Aphezulu Okukhiqizayo Kwezokuthutha )	- Ukulethwa okuzishayelayo kakhulu (amaloli, ama-drone) kuqondiswa abalawuli be-AI.- I-AI ishintsha ngokuzenzakalelayo ukuthunyelwa okuzungezile ukuphazamiseka futhi ilungisa isitokwe.- Ukuhlanganiswa kochungechunge lokuphakelwa kusukela ekuqaleni kuya ekugcineni (uku-oda, ukusatshalaliswa) okulawulwa yi-AI.
Ezezimali	- I-AI ikhiqiza imibiko yezezimali/izifinyezo zezindaba (ezibuyekezwe ngabantu).- Abeluleki be-Robo baphatha amaphothifoliyo alula; ingxoxo ye-AI iphatha imibuzo yamakhasimende. ( I-AI ekhiqizayo iyeza ukuzoxhasa )	- Abahlaziyi be-AI bakhiqiza izincomo zokutshalwa kwezimali kanye nemibiko yezingozi ngokunemba okuphezulu. - Ukulinganisa kabusha ukuhweba okuzimele kanye nephothifoliyo ngaphakathi kwemikhawulo ebekiwe. - I-AI ivuma ngokuzenzakalela imali mboleko/izimangalo ezijwayelekile; abantu babhekana nokungafaneleki.

Izinkomba:

1. EPatterson, ePhilana. Izindaba zemali engenayo ezizenzakalelayo ziyanda . I-Associated Press (2015) – Ichaza isizukulwane se-AP esizenzakalelayo sezinkulungwane zemibiko yemali engenayo ngaphandle kombhali ongumuntu ( Izindaba zemali engenayo ezizenzakalelayo ziyanda | I-Associated Press ).

2. I-McKinsey & Company. Isimo se-AI ekuqaleni kuka-2024: Ukwamukelwa kwe-Gen AI kuyanda futhi kuqala ukukhiqiza inani . (2024) - Ibika ama-65% ezinhlangano ezisebenzisa i-AI ekhiqizayo njalo, cishe kabili kusukela ngo-2023 ( Isimo se-AI ekuqaleni kuka-2024 | McKinsey ), futhi ixoxa ngemizamo yokunciphisa ubungozi ( Isimo se-AI: Ucwaningo Lomhlaba Wonke | McKinsey ).

3. UGartner. Ngale kwe-ChatGPT: Ikusasa le-AI Ekhiqizayo Yamabhizinisi . (2023) – Ibikezela ukuthi ngo-2030, ama-90% efilimu evelele kakhulu angase akhiqizwe yi-AI ( Ama-Generative AI Use Cases for Industries and Enterprises ) futhi igqamisa amacala okusetshenziswa kwe-AI ekhiqizayo njengokwakhiwa kwezidakamizwa ( Ama-Generative AI Use Cases for Industries and Enterprises ).

4. I-Twipe. Izindlela Eziyi-12 Izintatheli Ezisebenzisa Ngazo Amathuluzi E-AI Egumbini Lezindaba . (2024) – Isibonelo se-“Klara” AI esikhungweni sezindaba esibhala u-11% wezihloko, lapho abahleli abangabantu bebukeza konke okuqukethwe kwe-AI ( Izindlela Eziyi-12 Izintatheli Ezisebenzisa Ngazo Amathuluzi E-AI Egumbini Lezindaba - I-Twipe ).

5. Izindaba ze-Amazon.com. I-Amazon ithuthukisa ulwazi lokubuyekezwa kwamakhasimende nge-AI . (2023) - Imemezela izifinyezo zokubuyekezwa ezikhiqizwe yi-AI emakhasini omkhiqizo ukusiza abathengi ( I-Amazon ithuthukisa ulwazi lokubuyekezwa kwamakhasimende nge-AI ).

6. I-Zendesk. Izibalo zesevisi yamakhasimende ze-AI ezingama-59 zango-2025 . (2023) – Kubonisa ukuthi izinhlangano ze-CX ezingaphezu kwezingxenye ezimbili kwezintathu zicabanga ukuthi i-AI ekhiqizayo izokwengeza “ukufudumala” enkonzweni ( 59 Izibalo zesevisi yamakhasimende ze-AI zango-2025 ) futhi ibikezela i-AI ku-100% wokuxhumana kwamakhasimende ekugcineni ( 59 Izibalo zesevisi yamakhasimende ze-AI zango-2025 ).

7. Ucwaningo lwe-Futurum kanye ne-SAS. Okuhlangenwe nakho ngo-2030: Ikusasa Lokuhlangenwe nakho Kwamakhasimende . (2019) - Ucwaningo luthole ukuthi imikhiqizo ilindele ukuthi cishe ama-69% ezinqumo ngesikhathi sokuzibandakanya kwamakhasimende zizokwenziwa yimishini ehlakaniphile ngo-2030 ( Ukuze kucatshangelwe kabusha ukushintshela ku-CX, Abamakethi Kumelwe Benze Lezi Zinto Ezi-2 ).

8. Idathaiku. Amacala Okusetshenziswa Kwe-AI Aphezulu Okukhiqizayo ku-Logistics . (2023) – Ichaza indlela i-GenAI ethuthukisa ngayo ukulayisha (inciphisa isikhala seloli esingenalutho esingu-~30%) ( Amacala Okusetshenziswa Kwe-AI Aphezulu Okukhiqizayo ku-Logistics ) futhi iveza izingozi zochungechunge lokuhlinzekwa ngokuskena izindaba.

9. Umagazini i-Visual Studio. I-GitHub Copilot Ihamba Phambili Umbiko Wocwaningo Ngabasizi Bekhodi ye-AI . (2024) – Izibikezelo zokuhlela amasu zikaGartner: ngo-2028, ama-90% abathuthukisi bebhizinisi bazosebenzisa abasizi bekhodi ye-AI (kusuka ku-14% ngo-2024) ( Umbiko Wocwaningo lwe-GitHub Copilot Ihamba Phambili Ngabasizi Bekhodi ye-AI -- Umagazini i-Visual Studio ).

10. Izindaba zeBloomberg. Sethula iBloombergGPT . (2023) – Imininingwane ngemodeli yepharamitha yeBloomberg engu-50B ehlose imisebenzi yezezimali, eyakhelwe ku-Terminal yokusekela imibuzo nezimpendulo kanye nokuhlaziya ( i-AI ekhiqizayo iyeza ukuzoxhasa ).

Izihloko ongase uthande ukuzifunda ngemva kwalesi:

🔗 Imisebenzi Engenakushintshwa yi-AI - Futhi Yimiphi Imisebenzi I-AI Ezoshintshwa Yiyo?
Umbono womhlaba wonke ngesimo semisebenzi esishintshayo, uhlola ukuthi yiziphi izindima eziphephile ekuphazamisekeni kwe-AI nokuthi yiziphi ezisengozini enkulu.

🔗 Ingabe i-AI ingayibikezela imakethe yamasheya?
Ukujula ngamakhono, imikhawulo, kanye nokucatshangelwa kokuziphatha kokusebenzisa i-AI ekubikezeleni imakethe yamasheya.

🔗 Ingasetshenziswa Kanjani i-AI Ekhiqizayo Ekuphepheni Kwe-Cyber?
Funda ukuthi i-AI ekhiqizayo isetshenziswa kanjani ukuvikela izinsongo ze-cyber, kusukela ekutholakaleni okungavamile kuya ekubunjweni kwezinsongo.