Ngemuva kokwenza amagagasi phakathi nezinyanga zokuqala zango-2023, abasebenzisi be-inthanethi baya ngokuya baxhumana nenkundla ye-OpenAI's ChatGPT - kungakhathaliseki ukuthi bayayazi noma cha. Iziqalisi ziya ngokuya ziqala imikhiqizo egoqelwe ku-ChatGPT nama-chatbots, futhi izinkampani eziningi nakakhulu ziyabona ukuthi kungenzeka kube khona ukusetshenziswa kwe-ChatGPT kwangaphakathi.

Futhi, njenge-Robinhood kanye nokuhweba mahhala kwentando yeningi kanye nokutshala imali, i-ChatGPT iyakwazi nokwenza intando yeningi ukufunda ngezezimali okubanzi ngokwenza izihloko eziyinkimbinkimbi ziqondwe kalula futhi zigayeke. Kodwa akunjalo - ngisho ochwepheshe bezezimali bangabona izinsiza eziningi ekusetshenzisweni kwe-ChatGPT ngamasu ukuthuthukisa ukuthengisa, ukuthuthukisa imisebenzi, noma ukwandisa ukuguquguquka kwezezimali.

Yini i-ChatGPT? Uhlolojikelele

Amamodeli wezilimi ezinkulu (LLMs) ngokuvamile, kanye ne-ChatGPT ikakhulukazi, zisebenzisa imithombo ejulile yedatha yokuqeqeshwa, okuqukethwe, kanye nolwazi ukuze kuphendulwe izimpendulo kokokufaka okukhiqizwa umsebenzisi. Nakuba izinqubo nezindlela zixaka kakhulu ngaphansi kwe-hood, i-ChatGPT iphendula ukwaziswa ngokudlulisa lokho “ekufundile” esikhathini esidlule, bese ikhiqiza uchungechunge lwamagama okungenzeka ngokwezibalo ngokulandelana. Uma konke kusebenza kahle, impendulo yakhelwe phezu kwesisekelo sedatha ehloliwe, eqinisekiswa kokuphumayo okugayekayo, okufinyelelekayo, nokuqondakala kalula.

Esinye sezimo ezihlukile zokusetshenziswa ze-ChatGPT kungenxa yobunikazi noma idatha yokuqeqeshwa eyingqayizivele. Nakuba “amamodeli ayisisekelo” e-ChatGPT eqeqeshwa emikhawulweni yedatha ye-inthanethi etholakala esidlangalaleni, abasebenzisi “bangaphakela” i-LLM ngokuseduze, ulwazi lobunikazi ukuze ithuthukise izimpendulo zayo phakathi kwamasegimenti athile. Isibonelo, uma ngisebenzisa iwebhusayithi yokuhlaziya isitoko kanye nokutshalwa kwezimali, ngingakwazi “ukuphakela” i-ChatGPT ukuhlaziya kwami ​​​​okudlule kanye nedatha yesitoko yamanje bese ngiyicela ukuthi ibike ukuthi yiziphi izinyathelo zempumelelo ezichazwe kangcono.

Isibonelo, uma izindatshana zami ezigcizelela ukuhamba kwemali kwamahhala kwenkampani ngamamethrikhi esabelo ngasinye bekuyizinkomba ezihamba phambili zokusebenza kwesitoko esikhathini esizayo, ngingakwazi ukunciphisa imizamo yami ukuze ngigxile kuleso sici sezitatimende zezimali. Ngokufanayo, uma i-ChatGPT isiza ukunquma ukuthi isilinganiso senani ukuya-kumaholo sithembeke kancane inkomba yokusebenza ekuhlaziyeni kwami, ngingehlisa ukuthembela kwami ​​esilinganisweni esithile.

Okunye, izimo ezincane zokusebenzisa i-niche zingabandakanya ukuhlaziya nokusetshenziswa kwemali kokugeleza kwemali, okuphakelwa amadokhumenti okukhokha abathengisi, ukubikezela okuphakelwa amamethrikhi okusebenza adlule, nokunye (qhubeka ufunda ukuze uthole uhlu oluphelele!).

Kunjalo, “udoti ungene, ukhiphe udoti” uyasebenza, futhi izimpendulo ze-ChatGPT zilungile kuphela njengolwazi oluqondisayo. Ngesikhathi esifanayo, i-ChatGPT nama-LLM afanayo ivame ukungakutsheli lapho “engazi” okuthile futhi bavele bakulungise – ukuthambekela okubizwa ngokuthi, ngokufanelekile, ukubona izinto ezingekho.

Noma yikuphi ukusetshenziswa noma okukhiphayo kwe-ChatGPT kufanele kusetshenziswe njengelinye lamathuluzi amaningi, kungakhathaliseki ukuthi isetshenziswa kuphi, kodwa ikakhulukazi kwezezimali. Veza, qinisekisa, futhi uhlole izimpendulo ze-ChatGPT ngaphambi kokuzisebenzisa njengengxenye yohlelo lwezezimali noma lokusebenza, futhi uzobe usendleleni yokujwayela isimo sezezimali esishintsha ngokushesha.

Izindlela Eziphezulu Eziyisi-8 Zokusebenzisa I-ChatGPT Kwezezimali

Ngaphambi kokusebenzisa i-ChatGPT ngemisebenzi yezezimali, qiniseka ukuthi kunezidingo ezimbalwa ezitholakala ngaphambili:

1. Qinisekisa kwakho ihamba phambili zicacile futhi zimfushane.
2. Qinisekisa ukuthi noma iyiphi idatha "ophakelayo" imodeli igunyaziwe ukuthi ikhishwe noma ukuthi izivumelwano zakho ezithile zokuphepha zigunyaza ukusetshenziswa kwe-LLM ngedatha yokuphathelene.
3. Qinisekisa ukuthi wena noma ithimba lakho ninokuqonda okuqinile ngokwanele kwezimiso zezimali eziyisisekelo - uma ungaqondi ukuthi kwenzekani, mancane amathuba okuba uthole ulwazi olubi oluphumayo.

Ukuze kufezeke lokho, lawa angu-8 ezigameko ezidume kakhulu zokusebenzisa i-ChatGPT amaqembu ezezimali kuhlu lwezindima azihlolile onyakeni odlule futhi atholakala ethela kakhulu.

1. Ukukhipha Imibiko Yezezimali

Ukufakwa kwe-SEC nemibiko yezezimali ngamakhulu wamakhasi ubude, ngokuvamile nge i-legalese eyinkimbinkimbi kanye ne-jargon yezezimali kulo lonke. Ezezimali noma amaqembu amabhuku ingasebenzisa i-ChatGPT ngezindlela ezihlukahlukene zokwenza imibiko yezezimali icace, okuhlanganisa ukuchaza imigomo ethile noma ukufingqa izigaba ezinde. I-ChatGPT ingamukela uhla lwamafomethi okufakwayo kwedatha, kuhlanganise nama-PDF nama-URL aqondile, ngakho-ke ukuphakela i-LLM lolu lwazi kulula.

Umlayezo wesibonelo: Chaza ngesiNgisi esicacile ukuthi amanothi aguqukayo angaba nomthelela kanjani kusakhiwo senhloko-dolobha yenkampani, umbono wokuboleka, futhi kuthinte intengo yokwabelana.

2. Hlola Ukutshalwa Kwezimali Kwamanje

Okusetshenziselwa kangcono abeluleki bezezimali, abahleli, nabanye ochwepheshe (noma abatshalizimali ngabanye), ungakwazi ukuphakela ama-prospectuse esikhwama se-ChatGPT, ukubamba, nolunye ulwazi ukuze uhlaziye ukutshalwa kwezimali kwakho kwamanje sekukonke futhi wandise iphothifoliyo yakho namasu. Le ChatGPT yesu lokuhlaziya izimali ingase ibonise inguquko ekuletheni imfundo yokutshala imali enengqondo kubatshalizimali abadayisayo.

Umlayezo wesibonelo: Hlaziya le mibiko emithathu yokubamba kwe-ETF bese ungitshela ukuthi iziphi izitoko ezingazodwana ezinokugqagqana okukhulu ngaphakathi kwephothifoliyo, zibalwa ngesisindo esiphelele.

3. Izinkathazo Zokugeleza Kwemali eyimodeli - IngxoxoGPT Yokuhlaziywa Kwezezimali

Ngokuphakela i-ChatGPT izitatimende zakho ezedlule zokungena nokuphuma kwemali nokunye idatha yezezimali, kanye nolwazi lwenkokhelo oluqondene nomthengisi, ungasebenzisa inkundla ukuze ukhiqize idatha esetshenziswa ekumodeleni ukuze uhlole okuhlukile futhi uthuthukise isu lakho.

Umlayezo wesibonelo: Nquma umthelela isaphulelo senkokhelo yangaphambi kwesikhathi esingu-3% esinawo ekugelezeni kwemali kwami ​​konyaka, kucatshangwa ukuthi okwenziwa yikhasimende okungaphezu kuka-$10,000 kunethuba elingu-50% lokwamukela isaphulelo kanye nemisebenzi engaphansi kuka-$10,000 ibe nethuba elingu-30% lesaphulelo.

4. Dala Inhlabamkhosi Yezindaba Zezezimali

Kungakhathaliseki ukuthi imboni yakho, izindaba zezomnotho nezezimali zinomthelela wangempela futhi osheshayo emqondweni wakho - kodwa bambalwa bethu abanesikhathi noma umkhawulokudonsa wokuzinikezela ekuqondeni okujulile, okuguquguqukayo kwazo zonke izici eziqhuba umnotho wethu. Ungasebenzisa i-ChatGPT ukuze ufingqe izindatshana zezindaba noma wenze isifinyezo sosuku lokugcina usebenzisa izincwadi zezindaba ezithunyelwe nge-imeyili.

Ungakwazi ngisho nokufaka umkhakha wakho othize noma imboni ukuze uthole ukuqonda ongase ukuphuthele, udale inzuzo yokuncintisana.

Umlayezo wesibonelo: Fingqa amaphuzu abaluleke kakhulu avela kulezi zincwadi zezindaba zezomnotho zamasonto onke ezinhlanu ngokugcizelela ukuthi zingathinta kanjani ukuma kwami ​​njengomnikazi webhizinisi [INDUSTRY].

5. Thuthukisa Ukuxhumana Kwabatshalizimali

Nakuba konke ukuxhumana okubhaliwe okuqondile kufanele kuvele kubabhali bangempela, bekhwalithi (uma kungenjalo wena), ungasebenzisa i-ChatGPT ukuze siza ukuhlela imicabango yakho futhi udale ukuxhumana kwabatshalizimali okugeleza kahle futhi okunengqondo ngokombono womtshali-zimali - ikakhulukazi uma amakhasimende akho ejwayele ukungakwazi kahle ngokwezimali kunabasebenzi bakho.

Umlayezo wesibonelo: Isikhwama sami sokutshala sibuyise u-XX% onyakeni odlule, silindele ukuvala amadili ka-XX onyakeni ozayo, futhi ukukhathazeka kwami ​​komnotho omkhulu kuhlanganisa u-X, Y, no-Z. Uhlaka lohlaka lwencwadi yekota eya kubatshalizimali bami, okuhlanganisa la maphuzu amakhulu engigcizelela .

6. Hlela idatha

Enye yezimfanelo ezikhazimulayo, ezivelele ze-ChatGPT iwukusiza abantu abavamile ukuthi bakhiqize ikhodi yokusetshenziswa okuhlukahlukene. Okukodwa okulula nokufaneleka kakhulu kochwepheshe bezezimali akuyona ikhodi impela - kodwa kuseduze. I-ChatGPT ingasiza ukukhiqiza amafomula angacacile futhi ayinkimbinkimbi e-Excel nezinkundla ezifanayo, ithuthukise eyakho ukusebenza kahle ngokuzenzakalelayo.

Umlayezo wesibonelo: Ngineminyaka eyi-100 yedatha yomlando yembuyiselo ye-S&P 500, inkomba yamasheya amancane, inkomba yesitoko samazwe ngamazwe, ukwehla kwamandla emali, namazinga enzalo akhona. Khiqiza amafomula e-Excel ukuze usize ukunquma ukuhlobana phakathi kwamasethi edatha futhi uhlonze iminyaka lapho ukuhlobana bekuqine kakhulu. Bese, bhala isampula yokwabiwa kwempahla ekhulisa imbuyiselo yetiyori kuso sonke isikhathi.

7. Cwaninga Izimakethe Ezintsha noma Amathuba

Ngezindlela eziningi, i-ChatGPT ifana nenjini yokusesha kuma-steroids. Esikhundleni sokusesha amatemu nemishwana bese ifuna impendulo, i-ChatGPT ingagaya umbuzo cishe nganoma iyiphi ifomethi noma isitayela sokubhala futhi inikeze impendulo (efanele) ehambisanayo engenzeka uma ihlolwe kahle. Lokhu kuvula umhlaba wamathuba okukhipha imibono noma ukucwaninga amathuba amasha.

Umlayezo wesibonelo: Ngiyi-CFO yobuchwepheshe bokunakekelwa kwempilo ehlola amathuba amazwe ngamazwe. Imaphi amazwe anikezela ngezinzuzo ezinkulu zentela ezinkampanini ze-MedTech zokukhiqiza ngaphandle kolwandle?

8. Ukufingqa Imihlangano Nezingcingo

Sonke siyazi ukuthi ukubuyekezwa kokutshalwa kwezimali kwehora lonke kungenza amaklayenti athi shwaka, kodwa ukusebenzisa okulotshiweyo kocingo lwekota noma zonke izandla nge-ChatGPT kungaba indlela esheshayo yokunquma okuthathwayo okuphezulu, kanye nezinqumo zokuthwebula noma ukuphela kwesikhathi.

Umlayezo wesibonelo: Khiqiza isifinyezo esiphezulu samaphuzu avelele alolu cingo lokuzikhandla oluzothunyelwa ku-CFO. Dala futhi uhlu lwakho konke okulindelwe kusukela ku-port-co eduze nezinsuku ezilindelekile zokungabaza kanye nendawo yokuxhumana.

Imikhawulo kanye Nezinselele Zokusebenzisa I-ChatGPT Kwezezimali

Ngaphandle kwemibono eshiwo ngenhla kanye nokuphuma okungamanga ngezikhathi ezithile, ezinye izinselele ezihlukile zikhawulela ukusetshenziswa kwezimali kwe-ChatGPT kwabanye abasebenzisi:

• Kukhona ukubambezeleka kwedatha, futhi eminye imicimbi yakamuva ingase ingabikho kudatha yokuqeqeshwa ye-LLM.
• Ukufakwa kokwaziswa okufanele kudinga inhlawulo ethile okunzima ukuyibala uma ingekho.
• Kunengozi yokuputshuka kwedatha noma ebucayini, ikakhulukazi uma ulayisha ulwazi olubucayi noma lokuphathelene.

Isiphetho

 Nakuba kudinga ijika elincane lokufunda futhi ukuzimisela ukuzivumelanisa nezimo, I-ChatGPT kwezezimali ingase ikhulise ukusebenza kahle kwakho noma ithuthukise ukusebenza kwakho. Nakuba inezihibe ezithile futhi kungekona ukuphela konke lokho abanye abathi, njenganoma yiliphi ithuluzi i-ChatGPT enamacala okusebenzisa ahlukile, izinzuzo zayo eziningi zincike kusisebenzi esishwiba isando.

• I-SEO Powered Content & PR Distribution. Khuliswa Namuhla.
• I-PlatoData.Network Vertical Generative Ai. Zinike Amandla. Finyelela Lapha.
• I-PlatoAiStream. I-Web3 Intelligence. Ulwazi Lukhulisiwe. Finyelela Lapha.
• I-PlatoESG. Ikhabhoni, I-CleanTech, Amandla, Environment, Ilanga, Ukuphathwa Kwemfucuza. Finyelela Lapha.
• I-PlatoHealth. I-Biotech kanye ne-Clinical Trials Intelligence. Finyelela Lapha.
• Source: https://nanonets.com/blog/chatgpt-for-finance/