Umjaho Wezikhali we-AI ku-Cybersecurity: Ungathembeli Lutho, Qinisekisa Konke

Impi phakathi kwezigebengu ze-inthanethi nezingcweti ze-cybersecurity iyanda nsuku zonke. Umbiko we-cybersecurity uthi ubugebengu bamakhompuyutha bungabiza imali engango $10.5 trillion minyaka yonke ngo-2025. I-Artificial intelligence isibe isikhali kulo mjaho wezikhali. Njengo Unjiniyela Omkhulu Wokwethenjwa Kwesayithi (SRE) kwa-Microsoft nginolwazi lwe-cybersecurity lweminyaka engaphezu kwe-15, ngibone ukuvela kokuhlaselwa ku-inthanethi. Ngiphinde ngakubona ukubaluleka kwe-AI kwezokuphepha okuhambisana neziqu ze-Master's ezivela e-Georgia Tech kanye nesitifiketi se-CISSP; lokhu kunginike ulwazi olusebenzayo lokuthi i-AI isiza kanjani ekuvikelekeni kwe-inthanethi.

**I-AI-Powered Cyber Threats \ Izigebengu ze-Cybercriminal zisebenzisa i-AI ukuthuthukisa ukuhlasela okuyinkimbinkimbi nge-malware nobugebengu bokweba imininingwane ebucayi. Le mikhankaso yobugebengu bokweba imininingwane ebucayi isebenzisa i-AI ukuze ikholise abasebenzisi ukuthi baveze ulwazi olubucayi olufana namagama-mfihlo kanye nemininingwane yasebhange. I-AI isebenzisa ubuchwepheshe be-deepfake ukuzenza ubunikazi. Ngokwesibonelo, isisebenzi sezezimali sakhohliswa enkampanini yamazwe ngamazwe ukuba ikhokhe amaRandi ayizigidi ezingu-25 kubakhwabanisi kusetshenziswa ubuchwepheshe obungamanga. Abakhwabanisi bazenze abenkampani isikhulu esiphezulu sezezimali ocingweni lwengqungquthela yevidiyo, ngokusho kwamaphoyisa aseHong Kong.

Amaqembu obuhlakani obusabisayo kanye nezigameko manje abhekene nokuhlasela okusheshayo, okuqondiswe kakhulu okunzima ukukubona kusetshenziswa izindlela zokuphepha zendabuko.

Izigebengu ze-Cybercriminal zisebenzisa i-AI chatbots ukuzenza ukusekelwa kwamakhasimende. Kulokhu, amakhasimende azimisele ukwabelana ngedatha ebucayi njengamaphasiwedi nemininingwane yasebhange ngaphandle komcabango wesibili. Ngalokhu, izigebengu zifinyelela ulwazi lwebhange lwamakhasimende futhi zintshontshe kuma-akhawunti.

Ngaphezu kwalokho, i-AI isiza izigebengu ze-inthanethi ukuthi ziklame uhlelo olungayilungele ikhompyutha onjiniyela abangakazitholi, okwenza ukuba sengozini kwezinsuku eziyiziro kube inselele kakhulu.

Endimeni yami kumazisi nesizinda sezokuphepha se-Microsoft, ngibonile ukuthi amamodeli wokuthatha izinqumo aqhutshwa yi-AI asetshenziswa kanjani ukulwa nalezi zinsongo eziguqukayo.

I-AI njengebutho Lokuzivikela

Ngasohlangothini lokuzivikela, i-AI isiza i-cybersecurity. Amathuluzi okuphepha e-AI angahlaziya inani elikhulu ledatha, athole okudidayo, futhi enze izimpendulo ezizenzakalelayo kuzinsongo ezingaba khona. I-Zero Trust Architecture, uhlaka engilumelelayo ekuvikelekeni kwamafu, luthembele kakhulu ku-AI. Ithuluzi le-AI lihlala libuyekeza ukufinyelela komsebenzisi futhi liqinisekisa imininingwane. Lokhu kusiza ukunciphisa ukuhlaselwa kwama-akhawunti omsebenzisi nolwazi olubucayi. Ikhono le-AI lokubona ukuyekethisa liyenze yaba ithuluzi elibalulekile ku-DevSecOps kanye ne-automation. Ngenxa yalokhu, ukuphepha nokwethembeka kwesistimu kuye kwaba ngcono kakhulu.

Amathuluzi e-Artificial Intelligence angaxwayisa abalawuli nabasebenzisi ngokugebenga okungaba khona kanye nokuyekethisa. I-Microsoft inikeze amathuluzi e-AI abanzi nge-Identity and Access Management (IAM) ukuze kuqinisekiswe.

Ngaphandle kwalokhu, ochwepheshe be-cybersecurity kumele baqaphe futhi bahlole izexwayiso ezikhiqizwe ubunikazi kanye nethuluzi lokulawula ukufinyelela.

Iqhaza Le-Zero Trust kanye Nobunikazi Nokuphathwa Kokufinyelela (IAM)

I-mantra yeMicrosoft ye Zero Trust ithi “Ungalokothi uthembe, qinisekisa njalo”. Lokhu kudinga ukuqinisekisa ngokuqhubekayo abasebenzisi, ama-akhawunti, amadivayisi, nobunikazi. Izigebengu ze-Cybercriminal zifuna amafomu ahlukahlukene ukuzenza abasebenzisi, ngakho ukufakazela ubuqiniso kubalulekile, ikakhulukazi phakathi kwesizinda sikamazisi se-Microsoft.

I-Microsoft Identity and Access Management (IAM) iqinisekisa bonke ubuwena, i-akhawunti, kanye nokungena ngemvume kumodeli ye-zero trust. Lokhu kuqinisekisa ukuthi ubunikazi namadivayisi agunyaziwe ngaphambi kokunikeza ukufinyelela. Lokhu kwenziwa kusetshenziswa izindlela zokuqinisekisa ezifana ne-Multi-Factor Authentication (MFA) kanye nezinqubomgomo zokufinyelela ezinemibandela. Inkundla iMicrosoft eyisebenzisela ubunikazi nokulawula ukufinyelela ibizwa I-Microsoft Entra . I-Microsoft iphinde isebenzise i-Purview ne-Compliance ukuze ilawule ukuphepha. I-Defender isetshenziselwa ukuvikela amadivayisi, ama-akhawunti wamafu, nobunikazi. Abavikeli manje basebenzisa usongo olunamandla e-AI Ubuhlakani kanye Nempendulo Yesigameko, i-zero trust architecture, kanye ne-automation ukuze baphumelele ekutholeni nasekuphenduleni ngesikhathi sangempela.

I-Microsoft Entra isiza ngobuhlakani bokusongela kanye nempendulo yesigameko. Lokhu kusiza ekwehliseni ukuhlaselwa kanye nezinsongo zokuphepha. Ngempendulo yesigameko, ochwepheshe be-cybersecurity bangathola izexwayiso.

Indawo lapho iMicrosoft yenza kahle khona i-AI Threat Intelligence Platform. Ngo-2024, iMicrosoft yamemezela ukutshalwa kwezimali kwayo ku-AI Security ngokumema ochwepheshe ukuthi babambe iqhaza ku-Zero Day Quest ngezithembiso zemiklomelo yemali eningi. Lokhu kwenza iMicrosoft ibe ngumtshali zimali omkhulu ku-cybersecurity. I-Microsoft iphinda ihlaziye izigidigidi zamasiginali we-cybersecurity futhi inikeze amathuluzi ezinhlangano ezizowasebenzisa ekuzingeleni izinsongo.

Lobu buchwepheshe nezinhlelo zisiza ukwakha ukusimama nokuthembeka kwesistimu kochwepheshe.

I-DevSecOps: Ukuphepha ngesivinini sokuthuthukiswa

I-DevSecOps ihlanganise i-AI ku-cybersecurity, ukuphepha kwamafu, nokugeleza komsebenzi. I-DevSecOps iqinisekisa ukuthi ukuphepha kwakhelwa kuso sonke isigaba somjikelezo wempilo wokuthuthukiswa kwesofthiwe kunokuba kube umcabango wakamuva. Njengomamukeli we-Microsoft Award kanye nowine Indondo Yomxhumanisi Ongcono Kakhulu ngo-2016, ngibone izinhlangano zidonsa kanzima ngenxa yezinkinga zokuphepha. Ukuzenzakalela kwe-DevSecOps okunikwe amandla yi-AI kususa lezi zivimbamgwaqo, okuvumela izinkampani ukuthi zigcine isivinini nokuphepha.

I-AI-Powered Threat Intelligence & Incident Response

I-AI inika amandla ubuhlakani obusongelayo bokuthola ama-cyber hacks. Amamodeli e-AI asebenzisa ukufunda komshini ukuze ahlaziye amaphethini futhi athole okudidayo. Ngokufunda komshini, i-AI ibikezela izinsongo nezigameko.

Endimeni yami, ngisebenze ne-AI ezinhlelweni zokuphepha ezithola futhi ziphendule izinsongo ngokushesha kunezindlela zendabuko. Lokhu kutholwa kusize amaqembu okuvikela ukuthi asebenze futhi athole ngokushesha umsuka wezinkinga. Isistimu yezobunhloli obusabisayo ye-Microsoft ithola futhi inciphise ukugebenga, inikeza izexwayiso, futhi ivimbele ukungena ngemvume.

Ikusasa: Impi Eqhubekayo

Njengelungu Lebhodi Lokuhlela le ESP - Ijenali Yamazwe Ngamazwe Yentuthuko Yezobuchwepheshe Bekhompyutha kanye neLungu Eliphezulu le-IEEE, ngithole ithuba lokucwaninga nokuhlaziya intuthuko ye-AI kwezokuphepha. Impumelelo ku-cybersecurity incike ekuhlanganisweni kwe-AI, i-Zero Trust Architecture, i-IAM, i-DevSecOps automation, nochwepheshe abanamakhono.

Izigebengu ze-Cybercriminal zihlala zishintsha amaqhinga, futhi ochwepheshe be-cybersecurity kufanele bahlale behamba phambili ngamasu okuphepha athuthukisiwe we-AI. Into eyodwa eqinisekile, umjaho wezikhali we-AI ekuvikelekeni kwe-inthanethi usekude ukuphela. Kodwa-ke, nge-AI, kunamathuba angama-75% okuhlala ngaphambi kwezigebengu ze-inthanethi.