Ingabe i-AI ingathatha indawo ye-Cybersecurity?

Ingabe i-AI ingathatha indawo ye-Cybersecurity?

Impendulo emfushane: I-AI ngeke ithathe indawo yokuphepha kwe-cyber kusukela ekuqaleni kuya ekugcineni, kodwa izothatha izingxenye ezinkulu zomsebenzi wobunjiniyela be-SOC ophindaphindayo kanye nokuphepha. Isetshenziswa njengesinciphisi somsindo kanye nesifinyezo - nge-override yomuntu - isheshisa i-triage kanye nokubekwa phambili; iphathwa njenge-oracle, ingaletha ukuqiniseka okungamanga okuyingozi.

Izinto ezibalulekile okufanele uzicabangele:

Ububanzi : I-AI ithatha indawo yemisebenzi kanye nezindlela zokusebenza, hhayi umsebenzi ngokwawo noma ukuziphendulela.

Ukunciphisa umsebenzi : Sebenzisa i-AI ekuhlanganiseni izexwayiso, izifinyezo ezimfushane, kanye nokuhlola amaphethini e-log.

Ubunikazi bezinqumo : Gcina abantu bekulungele isifiso sokufaka engcupheni, umyalo wesigameko, kanye nokushintshana okunzima.

Ukumelana nokusetshenziswa kabi : Idizayini yokujova ngokushesha, ubuthi, kanye nemizamo yokugwema ukuphikisana.

Ukubusa : Qinisekisa imingcele yedatha, ukuhlolwa, kanye nokuguqulwa kwabantu okungaba khona ekusebenziseni amathuluzi.

Ingabe i-AI ingayithatha indawo ye-infographic yokuphepha kwe-cyber

Izihloko ongase uthande ukuzifunda ngemva kwalesi:

🔗 Indlela i-AI ekhiqizayo esetshenziswa ngayo ekuphepheni kwe-cyber
Izindlela ezisebenzayo ze-AI eziqinisa ukutholakala, impendulo, kanye nokuvimbela izinsongo.

🔗 Amathuluzi okufaka i-AI okuphepha kwe-cyber
Izixazululo eziphezulu ezisebenzisa i-AI zokuhlola ngokuzenzakalela nokuthola ubuthakathaka.

🔗 Ingabe i-AI iyingozi? Izingozi namaqiniso
Bheka ngokucacile izinsongo, izinganekwane, kanye nezivikelo ze-AI ezithembekile.

🔗 Umhlahlandlela wamathuluzi okuphepha e-AI aphezulu
Amathuluzi okuphepha angcono kakhulu asebenzisa i-AI ukuvikela izinhlelo nedatha.

Uhlaka "lokufaka esikhundleni" luyisicupho 😅

Uma abantu bethi “Ingabe i-AI ingathatha indawo ye-Cybersecurity” , bavame ukusho enye yezinto ezintathu:

  • Shintsha abahlaziyi (akudingeki abantu)

  • Shintsha amathuluzi (ipulatifomu eyodwa ye-AI yenza konke)

  • Buyisela imiphumela esikhundleni sayo (ukwephulwa okumbalwa, ingozi encane)

I-AI inamandla kakhulu ekuthatheni indawo yemizamo ephindaphindwayo kanye nokucindezela isikhathi sokwenza izinqumo. Ibuthakathaka kakhulu ekuthatheni indawo yokuziphendulela, umongo, kanye nokwahlulela. Ukuphepha akukhona nje ukutholakala - kuyizingxabano ezinzima, imingcele yebhizinisi, ezombusazwe (ugh), kanye nokuziphatha kwabantu.

Uyazi ukuthi kuhamba kanjani - ukwephulwa kwemithetho kwakungekona “ukuntuleka kwezixwayiso.” Kwakuwukuntuleka komuntu okholelwa ukuthi lesi saziso sibalulekile. 🙃

Lapho i-AI isivele “ithatha indawo” yomsebenzi wokuphepha kwe-cyber (ngokoqobo) ⚙️

I-AI isivele ithatha izigaba ezithile zomsebenzi, noma ngabe ishadi lenhlangano lisabukeka lifana.

1) Ukuhlelwa kwe-triage kanye nokuhlanganiswa kwezexwayiso

  • Ukuhlanganisa izexwayiso ezifanayo zibe yisigameko esisodwa

  • Ukususa ukukopisha izimpawu ezinomsindo

  • Isilinganiso ngomthelela ongaba khona

Lokhu kubalulekile ngoba i-triage yilapho abantu belahlekelwa khona yintando yabo yokuphila. Uma i-AI inciphisa umsindo ngisho nakancane, kufana nokuvala i-alamu yomlilo ebilokhu ikhala amasonto amaningi 🔥🔕

2) Ukuhlaziywa kwelogi kanye nokutholwa kwe-anomaly

  • Ukubona amaphethini asolisayo ngesivinini somshini

  • Ukumaka "lokhu akujwayelekile uma kuqhathaniswa nesisekelo"

Ayiphelele, kodwa ingaba wusizo. I-AI ifana ne-metal detector ebhishi - ikhala kakhulu, futhi ngezinye izikhathi iyisivalo sebhodlela, kodwa ngezinye izikhathi iyindandatho 💍… noma ithokheni yokuphatha ethintekile.

3) Ukuhlukaniswa kwe-Malware kanye ne-phishing

  • Ukuhlukanisa okunamathiselwe, ama-URL, izizinda

  • Ukuthola izinhlobo ezifanayo kanye namaphethini okukhohlisa

  • Izifinyezo zesinqumo se-sandbox ezenzakalelayo

4) Ukubekwa phambili kokuphathwa kobungozi

Akusikho ukuthi “yimaphi ama-CVE akhona” - sonke siyazi ukuthi maningi kakhulu. I-AI iyasiza ekuphenduleni:

Futhi yebo, abantu bangakwenza lokho - ukube isikhathi besingenamkhawulo futhi akekho owake wathatha amaholide.

Yini eyenza inguqulo ye-AI ibe yinhle ekuphepheni kwe-cyber 🧠

Lena ingxenye abantu abayigwemayo, bese besola “i-AI” njengokungathi umkhiqizo owodwa onemizwa.

Inguqulo enhle ye-AI ekuphepheni kwe-cyber ivame ukuba nalezi zici:

  • Ukuqeqeshwa okuphezulu kwesignali kuya kumsindo

    • Kumelwe kunciphise umsindo, hhayi ukukhulisa umsindo owengeziwe ngamagama amahle.

  • Ukuchaza okusiza ekusebenzeni

    • Akuyona inoveli. Akuyona imizwa. Izinkomba zangempela: lokho ekubonile, ukuthi kungani ikhathalela, yini eshintshile.

  • Ukuhlanganiswa okuqinile nendawo okuyo

    • I-IAM, i-endpoint telemetry, ukuma kwefu, amathikithi, isitokwe sempahla… izinto ezingathandeki.

  • Ukufakwa kwe-overhide komuntu okwakhelwe ngaphakathi

    • Abahlaziyi kudingeka bakulungise, bakulungise, futhi ngezinye izikhathi bakushaye indiva. Njengomhlaziyi omncane ongalali kodwa ngezinye izikhathi othukayo.

  • Ukuphathwa kwedatha okuphephile

    • Sula imingcele yalokho okugcinwayo, okuqeqeshwayo, noma okugcinwayo. I-NIST AI RMF 1.0

  • Ukuqina ngokumelene nokuxhashazwa

Masikhulume iqiniso - "ukuphepha kwe-AI" okuningi kuyahluleka ngoba kuqeqeshwe ukuzwakala kuqinisekile, hhayi ukuthi kulungile. Ukuzethemba akuyona into yokulawula. 😵💫

Izingxenye ze-AI ziyahluleka ukuzishintsha - futhi zibaluleke kakhulu kunalokho ezikuzwakalayo 🧩

Nansi iqiniso elingajabulisi: ukuphepha kwe-inthanethi akugcini nje ngobuchwepheshe. Kungokwezenhlalo-zobuchwepheshe. Kungabantu kanye nezinhlelo kanye nezikhuthazo.

I-AI ilwa no:

1) Umongo webhizinisi kanye nesifiso sobungozi

Izinqumo zokuphepha azivamile ukuthi “zimbi yini.” Zifana kakhulu nalezi:

  • Ukuthi kukhulu ngokwanele yini ukuvimba imali engenayo

  • Kungakhathaliseki ukuthi kufanelekile yini ukuphula ipayipi lokuthunyelwa

  • Ingabe ithimba eliphethe lizokwamukela isikhathi sokuphumula salokho?

I-AI ingasiza, kodwa ayikwazi ukuba yikho lokho. Othile usayina igama lakhe esinqumweni. Othile uthola ucingo ngo-2 ekuseni 📞

2) Umyalo wesigameko kanye nokuxhumana kweqembu lonke

Ngesikhathi sezigameko zangempela, "umsebenzi" uthi:

I-AI ingabhala umugqa wesikhathi noma ifingqe izingodo, yebo. Ukufaka esikhundleni sobuholi ngaphansi kwengcindezi... kunethemba. Kufana nokucela umshini wokubala ukuthi usebenzise i-fire drill.

3) Ukumodela okusongelayo kanye nokwakhiwa kwezakhiwo

Ukumodela okusongelayo kuyingxenye yokucabanga, ingxenye yokudala, ingxenye yokuphambana (uphambana okunempilo, ikakhulukazi).

  • Ukubala ukuthi yini engase ihambe kabi

  • Ukulindela lokho umhlaseli angakwenza

  • Ukukhetha ukulawula okushibhile okushintsha izibalo zomhlaseli

I-AI ingaphakamisa amaphethini, kodwa inani langempela livela ekwazini izinhlelo zakho, abantu bakho, izinqamuleli zakho, kanye nokuxhomekeka kwakho okungavamile kwefa.

4) Izici zabantu kanye namasiko

Ubugebengu bokweba imininingwane ebucayi, ukusetshenziswa kabusha kweziqinisekiso, i-shadow IT, ukubuyekezwa kokufinyelela okungahlelekile - lezi yizinkinga zabantu abagqoke izingubo zobuchwepheshe 🎭
I-AI ingayibona, kodwa ayikwazi ukulungisa ukuthi kungani inhlangano iziphatha ngendlela eziphatha ngayo.

Abahlaseli basebenzisa i-AI - ngakho inkundla yokudlala ithambekela eceleni 😈🤖

Noma iyiphi ingxoxo yokushintsha ukuphepha kwe-inthanethi kufanele ifake okusobala: abahlaseli abami ndawonye.

I-AI isiza abahlaseli:

Ngakho-ke abavikeli abasebenzisa i-AI akuyona into ongayikhetha isikhathi eside. Kufana nokuthi… uletha ithoshi ngoba olunye uhlangothi lusanda kuthola izibuko zokubona ebusuku. Isingathekiso esingacacile. Kuseyiqiniso.

Futhi, abahlaseli bazohlasela izinhlelo ze-AI ngokwabo:

Ukuphepha bekulokhu kufana nekati negundane. I-AI yenza amakati asheshe futhi amagundane abe nobuciko obusha 🐭

Impendulo yangempela: I-AI ithatha indawo yemisebenzi, hhayi ukuziphendulela ✅

Leli yiqembu “eliphakathi nendawo elibi” iningi lalo elifika kulo:

  • I-AI iphatha isikali

  • Abantu baphatha izigxobo

  • Ndawonye baphatha isivinini kanye nokwahlulela

Ekuhlolweni kwami ​​​​kokusebenza kokuphepha, i-AI ingcono kakhulu uma iphathwa kanje:

  • Umsizi wokuhlola

  • Isifinyezo

  • Injini yokuhlanganisa

  • Umsizi wenqubomgomo

  • Umngane wokubuyekezwa kwekhodi ngamaphethini ayingozi

I-AI iba yimbi kakhulu uma iphathwa kanje:

  • Isazi sezinkanyezi

  • Iphuzu elilodwa leqiniso

  • Uhlelo lokuzivikela "olusethe bese ulukhohlwa"

  • Isizathu sokungasebenzi kahle kweqembu (leli liluma kamuva... kakhulu)

Kufana nokuqasha inja eqaphayo ebhala nama-imeyili. Kuhle. Kodwa ngezinye izikhathi ikhonkotha endaweni engenalutho bese iphuthelwa yindoda egxuma ocingweni. 🐶🧹

Ithebula Lokuqhathanisa (izinketho eziphezulu amaqembu azisebenzisayo nsuku zonke) 📊

Ngezansi kunethebula lokuqhathanisa elisebenzayo - aliphelele, alilingani kancane, njengempilo yangempela.

Ithuluzi / Ipulatifomu Kuhle kakhulu (kwezithameli) Isimo sentengo Kungani kusebenza (kanye nezimpawu ezingavamile)
I-Microsoft Sentinel I -Microsoft Learn Amaqembu e-SOC ahlala ezindaweni zemvelo ze-Microsoft $$ - $$$ Amaphethini e-SIEM aqinile avela efwini; izixhumi eziningi, zingaba nomsindo uma zingalungiswanga…
Ukuphepha Kwebhizinisi Okubi Kakhulu Ama-org amakhulu anokuqopha okuningi + izidingo ezenziwe ngokwezifiso $$$ (ngokuvamile $$$$ ngokukhuluma iqiniso) Ukusesha okunamandla + amadeshibhodi; kuyamangalisa uma kukhethwe kahle, kubuhlungu lapho kungekho muntu onenhlanzeko yedatha
Imisebenzi Yokuphepha kwe-Google Ifu le-Google Amaqembu afuna i-telemetry elawulwayo $$ - $$$ Kuhle kakhulu ku-big data scale; kuncike ekuvuthweni kokuhlanganiswa, njengezinto eziningi
I-CrowdStrike Falcon CrowdStrike Amaqembu e-Endpoint-heavy, amaqembu e-IR $$$ Ukubonakala okuqinile kwe-endpoint; ukujula okuhle kokuthola, kodwa usadinga abantu abazoqhuba impendulo
I-Microsoft Defender ye-Endpoint I-Microsoft Learn Ama-org anzima e-M365 $$ - $$$ Ukuhlanganiswa kweMicrosoft okuqinile; kungaba kuhle kakhulu, kungaba “izaziso ezingu-700 emgqeni” uma kungalungiselelwanga kahle
Amanethiwekhi e- Palo Alto Cortex XSOAR Ama-SOC agxile ezenzakalelayo $$$ Izincwadi zokudlala zinciphisa umsebenzi; zidinga ukunakekelwa noma uzenzakalela ukuphazamiseka (yebo lokho kuyinto)
Ipulatifomu ye-Wiz Amaqembu okuphepha kwamafu $$$ Ukubonakala okuqinile kwamafu; kusiza ukubeka phambili ingozi ngokushesha, kusadinga ukuphathwa kwayo
Ipulatifomu ye-Snyk Ama-org okuqala onjiniyela, i-AppSec $$ - $$$ Izindlela zokusebenza ezilungele abathuthukisi; impumelelo incike ekwamukelweni konjiniyela, hhayi ukuskena nje kuphela

Inothi elincane: akukho thuluzi "eliwina" lodwa. Ithuluzi elihle kakhulu yilelo iqembu lakho elilisebenzisa nsuku zonke ngaphandle kokulicasula. Akuyona isayensi leyo, ukusinda lokho 😅

Imodeli yokusebenza engokoqobo: indlela amaqembu anqoba ngayo nge-AI 🤝

Uma ufuna i-AI ithuthukise ukuphepha ngendlela enenjongo, incwadi yokudlala ivame ukuba:

Isinyathelo 1: Sebenzisa i-AI ukunciphisa umsebenzi

  • Izifinyezo zokucebisa izexwayiso

  • Ukubhalwa kwamathikithi

  • Uhlu lokuhlola lokuqoqwa kobufakazi

  • Iziphakamiso zombuzo welogi

  • "Okwashintsha" kuyahluka kuma-config

Isinyathelo sesi-2: Sebenzisa abantu ukuqinisekisa nokunquma

  • Qinisekisa umthelela kanye nobubanzi

  • Khetha izenzo zokuvimbela

  • Ukulungisa izilungiso phakathi kwamaqembu ahlukahlukene

Isinyathelo 3: Yenza izinto eziphephile zibe ngokuzenzakalelayo

Izinhloso ezinhle zokwenza ngokuzenzakalelayo:

  • Ukuvalela amafayela aziwayo ukuthi amabi ngokuzethemba okukhulu

  • Ukusetha kabusha iziqinisekiso ngemuva kokuvumelana okuqinisekisiwe

  • Ukuvimba izizinda ezinobungozi ngokusobala

  • Ukuphoqelela ukulungiswa kokuzulazula kwenqubomgomo (ngokuqapha)

Izinhloso zokuzenzakalela eziyingozi:

  • Ukuhlukanisa ngokuzenzakalela amaseva okukhiqiza ngaphandle kwezivikelo

  • Ukususa izinsiza ngokusekelwe kumasignali angaqinisekile

  • Ukuvimba amabanga amakhulu e-IP ngoba "imodeli yayizizwa sengathi injalo" 😬

Isinyathelo 4: Buyisela izifundo ezilawulini

  • Ukulungiswa kwangemva kwesigameko

  • Ukutholwa okuthuthukisiwe

  • Isitoko sempahla esingcono (ubuhlungu obungapheli)

  • Amalungelo amancane

Yilapho i-AI isiza khona kakhulu: ukufingqa ukuhlolwa kwezidumbu, ukudweba izikhala zokuthola, ukuguqula ukuphazamiseka kube ngcono okuphindaphindwayo.

Izingozi ezifihliwe zokuphepha okuqhutshwa yi-AI (yebo, zimbalwa) ⚠️

Uma usebenzisa i-AI kakhulu, udinga ukuhlela izinto ezilandelayo:

Kufana nokwakha ilokhi ehlakaniphile kakhulu bese ushiya ukhiye ngaphansi komata. Ilokhi akuyona yodwa inkinga.

Ngakho-ke… Ingabe i-AI ingathatha indawo ye-Cybersecurity: impendulo ehlanzekile 🧼

Ingabe i-AI ingathatha indawo ye-Cybersecurity?
Ingathatha indawo yomsebenzi omningi ophindaphindwayo ngaphakathi kwe-cybersecurity. Ingasheshisa ukutholwa, ukuhlolwa kwezibalo, ukuhlaziywa, ngisho nezingxenye zempendulo. Kodwa ayikwazi ukuthatha indawo ngokugcwele yesiyalo ngoba i-cybersecurity akuyona umsebenzi owodwa - ukuphatha, ukwakheka, ukuziphatha kwabantu, ubuholi bezehlakalo, kanye nokuzivumelanisa nezimo okuqhubekayo.

Uma ufuna ukwakheka okuqondile kakhulu (okungacacile kancane, uxolo):

  • I-AI ithatha indawo yomsebenzi omatasa

  • I-AI ithuthukisa amaqembu amahle

  • I-AI idalula izinqubo ezimbi

  • Abantu bahlala benesibopho ngengozi kanye neqiniso

Futhi yebo, ezinye izindima zizoshintsha. Imisebenzi yezinga lokuqala izoshintsha ngokushesha. Kodwa imisebenzi emisha nayo iyavela: imisebenzi ephephile ngokushesha, ukuqinisekiswa kwemodeli, ubunjiniyela bokuzenzakalela kokuphepha, ubunjiniyela bokuthola ngamathuluzi asizwa yi-AI… umsebenzi awunyamalali, uyashintsha 🧬

Amanothi okuvala kanye nesifinyezo esisheshayo 🧾✨

Uma unquma ukuthi wenzeni nge-AI ngokuphepha, nansi into ewusizo ongayenza:

  • Sebenzisa i-AI ukucindezela isikhathi - ukulinganisa okusheshayo, izifinyezo ezisheshayo, ukuxhumanisa okusheshayo.

  • Gcina abantu benelungelo lokwahlulela - umongo, ukushintshisana, ubuholi, ukuziphendulela.

  • Ake sithi abahlaseli basebenzisa i-AI nabo - umklamo wokukhohlisa nokukhohlisa. ze-MITER ATLAS Zokuthuthukiswa Kwesistimu Ye-AI Evikelekile (NSA/CISA/NCSC-UK)

  • Ungathengi “umlingo” - thenga izindlela zokusebenza ezinciphisa ubungozi kanye nomsebenzi ngendlela elinganiselayo.

Ngakho-ke yebo, i-AI ingathatha indawo yezingxenye zomsebenzi, futhi ivame ukwenza kanjalo ngezindlela ezizwakala zingabonakali ekuqaleni. Isinyathelo esiphumelelayo ukwenza i-AI ibe yithuluzi lakho lokusiza, hhayi indawo yakho.

Futhi uma ukhathazekile ngomsebenzi wakho - gxila ezintweni i-AI elwa nazo: ukucabanga kwezinhlelo, ubuholi bezigameko, ukwakheka kwezakhiwo, kanye nokuba ngumuntu okwazi ukuhlukanisa phakathi "kwesexwayiso esithakazelisayo" kanye nokuthi "sizoba nosuku olubi kakhulu." 😄🔐

Imibuzo Evame Ukubuzwa

Ingabe i-AI ingathatha indawo ngokuphelele yamaqembu okuphepha kwe-cyber?

I-AI ingathatha izingxenye ezinkulu zomsebenzi wokuphepha kwe-cyber, kodwa hhayi isiyalo kusukela ekuqaleni kuze kube sekupheleni. Iphumelela kakhulu emisebenzini ephindaphindwayo yokuphuma njengokuqoqa izixwayiso, ukuthola okungahambi kahle, kanye nokubhala izifinyezo zokuphasa kokuqala. Okungayithathi indawo ukuziphendulela, umongo webhizinisi, kanye nokwahlulela lapho izingozi ziphezulu. Empeleni, amaqembu ahlala “esikhundleni esibi” lapho i-AI iletha khona isikali kanye nesivinini, kuyilapho abantu begcina ubunikazi bezinqumo ezilandelanayo.

Kuphi lapho i-AI isivele ithatha khona indawo yomsebenzi wansuku zonke we-SOC?

Kuma-SOC amaningi, i-AI isivele ithatha umsebenzi onzima njenge-triage, i-de-duplication, kanye nezexwayiso zokubeka amazinga ngokwemiphumela engaba khona. Ingasheshisa futhi ukuhlaziywa kwelogi ngokumaka amaphethini asuka ekuziphatheni okuyisisekelo. Umphumela awubi mbalwa kakhulu ngezigameko zomlingo - kunciphisa amahora achithwa udlula emsindweni, ngakho abahlaziyi bangagxila ophenyweni olubalulekile.

Amathuluzi e-AI asiza kanjani ngokuphathwa kobuthakathaka kanye nokubeka phambili ama-patch?

I-AI isiza ekushintsheni ukuphathwa kobuthakathaka kusuka "kuma-CVE amaningi kakhulu" kuya kokuthi "yini okufanele siyilungise kuqala lapha." Indlela evamile ihlanganisa izimpawu zokusebenziseka kalula (njenge-EPSS), uhlu lokuxhashazwa olwaziwayo (njengekhathalogi ye-CISA ye-KEV), kanye nomongo wendawo yakho (ukuvezwa kwe-inthanethi kanye nokubaluleka kwempahla). Kwenziwe kahle, lokhu kunciphisa ukuqagela futhi kusekela ukulungisa ngaphandle kokuphula ibhizinisi.

Yini eyenza i-AI "enhle" ekuphepheni kwe-cyber uma kuqhathaniswa ne-AI enomsindo?

I-AI enhle ekuphepheni kwe-inthanethi inciphisa umsindo kunokukhiqiza imfuhlumfuhlu ezwakala sengathi iyazethemba. Inikeza ukuchazeka okusebenzayo - izinkomba eziqondile ezifana nokuthi yini eshintshile, ukuthi yini eyibonile, nokuthi kungani ibalulekile - esikhundleni sezindaba ezinde nezingaqondakali. Iphinde ihlanganiswe nezinhlelo eziyinhloko (i-IAM, i-endpoint, ifu, ithikithi) futhi isekela ukuguqulwa kwabantu ukuze abahlaziyi bakwazi ukukulungisa, ukukulungisa, noma ukungakunaki lapho kudingeka.

Yiziphi izingxenye zokuphepha kwe-cyber i-AI ehluleka ukuzishintsha?

I-AI inenkinga enkulu ngomsebenzi wezenhlalo nobuchwepheshe: isifiso sobungozi, umyalo wesigameko, kanye nokuxhumana kwamaqembu amaningi. Ngesikhathi sezehlakalo, umsebenzi uvame ukuba ukuxhumana, ukuphathwa kobufakazi, ukukhathazeka kwezomthetho, kanye nokwenza izinqumo ngaphansi kokungaqiniseki - izindawo lapho ubuholi budlula khona ukufanisa amaphethini. I-AI ingasiza ekufinyezeni izingodo noma izikhathi zokuhlela, kodwa ayithathi indawo yobunikazi ngokuthembekile ngaphansi kwengcindezi.

Abahlaseli basebenzisa kanjani i-AI, futhi ingabe lokho kuyawushintsha umsebenzi womvikeli?

Abahlaseli basebenzisa i-AI ukwandisa i-phishing, bakhiqize ubunjiniyela bezenhlalo obukholisayo, futhi baphinde basebenzise izinhlobo ze-malware ngokushesha. Lokho kushintsha inkundla yokudlala: abavikeli abasebenzisa i-AI baba yinto engakhethwa kangako ngokuhamba kwesikhathi. Kunezela nengozi entsha, ngoba abahlaseli bangase bahlose imisebenzi ye-AI ngokujova ngokushesha, ukuzama ukufaka ubuthi, noma ukugwema ukuphikisana - okusho ukuthi izinhlelo ze-AI nazo zidinga ukulawulwa kokuphepha, hhayi ukuthembana okungaboni.

Yiziphi izingozi ezinkulu zokuthembela ku-AI ngezinqumo zokuphepha?

Ingozi enkulu ukuqiniseka okusunguliwe: I-AI ingazwakala iqinisekile ngisho noma ingalungile, futhi ukuzethemba akuyona into yokulawula. Ukuvuza kwedatha kungenye ingozi evamile - izixwayiso zokuphepha zingase zifaka imininingwane ebucayi ngengozi, futhi amalogi avame ukuqukatha izimfihlo. Ukuthembela ngokweqile kungaphinde kuphazamise izisekelo, kuyilapho ukuzulazula kwemodeli kunciphisa kancane ukutholwa njengoba izindawo nokuziphatha komhlaseli kushintsha.

Iyini imodeli yokusebenza engokoqobo yokusebenzisa i-AI ekuphepheni kwe-cyber?

Imodeli esebenzayo ibukeka kanje: sebenzisa i-AI ukunciphisa umsebenzi, ugcine abantu ukuze baqinisekise futhi benze izinqumo, futhi wenze izinto eziphephile zibe ngokuzenzakalelayo kuphela. I-AI inamandla ezifinyezweni zokucebisa, ukubhala amathikithi, uhlu lokuhlola lobufakazi, kanye nokwehluka "kwalokho okushintshile". Ukuzenzakalela kufaneleka kakhulu ezenzweni zokuzethemba okuphezulu njengokuvimba izizinda ezaziwayo ezimbi noma ukusetha kabusha iziqinisekiso ngemuva kokuvumelana okuqinisekisiwe, ngezivikelo zokuvimbela ukweqisa.

Ingabe i-AI izothatha indawo yezindima zokuphepha kwe-cyber ezingeni lokuqala, futhi yimaphi amakhono azoba yigugu kakhulu?

Imisebenzi eminingi ezingeni lokuqala cishe izoshintsha ngokushesha ngoba i-AI ingamunca ukuhlolwa okuphindaphindiwe, ukufingqa, kanye nomsebenzi wokuhlukanisa. Kodwa imisebenzi emisha nayo iyavela, njengokwakha imisebenzi ephephile ngokushesha, ukuqinisekisa imiphumela yamamodeli, kanye nokwenza ngokuzenzakalelayo kokuphepha kobunjiniyela. Ukuqina komsebenzi kuvame ukuvela kumakhono i-AI elwa nawo: ukucabanga kwezinhlelo, ukwakheka kwezakhiwo, ubuholi bezigameko, kanye nokuhumusha izimpawu zobuchwepheshe zibe izinqumo zebhizinisi.

Izinkomba

  1. OKUQALA - EPSS (OKUQALA) - first.org

  2. I-ejensi Yokuphepha Kwe-inthanethi kanye Nengqalasizinda (i-CISA) - Ikhathalogi Eyaziwayo Yokuxhashazwa Okuxhashazwayo - cisa.gov

  3. Isikhungo Sikazwelonke Sezindinganiso Nobuchwepheshe (i-NIST) - SP 800-40 Rev. 4 (Ukuphathwa Kwezingxenyana Zebhizinisi) - csrc.nist.gov

  4. Isikhungo Sikazwelonke Sezindinganiso Nobuchwepheshe (i-NIST) - i-AI RMF 1.0 - nvlpubs.nist.gov

  5. I-OWASP - LLM01: Ukujova Okusheshayo - genai.owasp.org

  6. Uhulumeni wase-UK - Imithetho yokusebenza yokuphepha kwe-inthanethi kwe-AI - gov.uk

  7. Isikhungo Sikazwelonke Sezindinganiso Nobuchwepheshe (i-NIST) - SP 800-61 (Umhlahlandlela Wokusingatha Izehlakalo) - csrc.nist.gov

  8. I-Federal Bureau of Investigation (FBI) - I-FBI ixwayisa ngokwanda kosongo lwezigebengu ze-cyber ezisebenzisa ubuhlakani bokwenziwa - fbi.gov

  9. Isikhungo Sezikhalazo Zobugebengu Be-inthanethi se-FBI (IC3) - I-IC3 PSA ngokukhwabanisa/ubugebengu bokweba imininingwane ebucayi be-AI - ic3.gov

  10. I-OpenAI - Imibiko yobunhloli bezinsongo ze-OpenAI (izibonelo zokusetshenziswa okunonya) - openai.com

  11. I-Europol - I-Europol “Umbiko we-ChatGPT” (ukubuka konke kokusetshenziswa kabi) - europol.europa.eu

  12. MITRE - I-MITRE ATLAS - mitre.org

  13. I-OWASP - I-OWASP Eziyi-10 Eziphezulu Zezicelo ze-LLM - owasp.org

  14. I-National Security Agency (NSA) - Isiqondiso Sokuvikela Ukuthuthukiswa Kwesistimu ye-AI (NSA/CISA/NCSC-UK kanye nabalingani) - nsa.gov

  15. I-Microsoft Learn - Ukubuka konke kwe-Microsoft Sentinel - learn.microsoft.com

  16. I-Splunk - Ukuphepha Kwebhizinisi Okubi - splunk.com

  17. I-Google Cloud - Imisebenzi Yokuphepha ye-Google - cloud.google.com

  18. I-CrowdStrike - Ipulatifomu ye-CrowdStrike Falcon - crowdstrike.com

  19. I-Microsoft Learn - Microsoft Defender ye-Endpoint - learn.microsoft.com

  20. Amanethiwekhi e-Palo Alto - i-Cortex XSOAR - paloaltonetworks.com

  21. I-Wiz - Ipulatifomu ye-Wiz - wiz.io

  22. I-Snyk - I-Snyk Platform - snyk.io

Thola i-AI Yakamuva Esitolo Esisemthethweni Somsizi we-AI

Mayelana NATHI

Buyela kubhulogi