Haɗin Haɗin Rashin Rauni na Zero-Ray da Ci gaban AI: Damuwa mai Girma
Masana'antar fasaha tana kewaya wani wuri mai rikitarwa inda rashin lahani na kwana-kwana da saurin ci gaban AI ke shiga tsakani, suna haifar da cikakkiyar guguwa wanda kwararrun tsaro dole ne su magance. Don fahimtar wannan batu, yana da mahimmanci a ayyana mahimman kalmomi: rashin lahani na kwana-kwana yana nufin kurakuran tsaro da ba a san su ba waɗanda maharan za su iya amfani da su a baya, yayin da ci gaban AI ya haɗa da haɓaka tsarin bayanan ɗan adam wanda zai iya aiwatar da ayyuka da kansa. Kamar yadda aka ruwaito [1], adadin rashin lahani na kwana-kwana da ake amfani da shi a cikin daji yana ƙaruwa, tare da fitattun misalan da suka haɗa da lahani mai mahimmanci da aka gano a cikin samfuran Check Point da Apple's iPhones da iPads. A halin yanzu, sararin AI yana shaida mahimman ƙira, gami da ƙaddamar da GPT-5.6 Sol da Grok 4.5, da kuma siyan Ona don Codex [2].
Yunkuri na baya-bayan nan na rashin lahani na kwana-kwana yana nuna karuwar barazanar hare-hare na kwana-kwana, wadanda ke amfani da raunin da ba a san su ba a baya, wanda ke barin kungiyoyin tsaro da karancin lokaci don mayar da martani. Misali, rashin lafiyar ranar sifili na Check Point, CVE-2026-16232, yana da sakamako mai tsanani, gami da daidaita bayanai masu mahimmanci da rushewar tsarin mahimmanci [12]. Gaskiyar cewa CISA ta ƙara wannan raunin zuwa ga Katalojin da aka sani da ake amfani da shi (KEV) [13] yana nuna tsananin barazanar. Don kwatanta tasirin hare-haren kwana-kwana, yi la’akari da batun ƙetare Equifax, inda rashin lahani na kwana-kwana a cikin software na Apache Struts ya haifar da fallasa mahimman bayanai ga mutane sama da miliyan 147.
Saurin saurin haɓakar AI shine takobi mai kaifi biyu. A gefe guda, AI yana da damar canza masana'antu da canza yadda muke rayuwa da aiki. A gefe guda, karuwar dogaro ga tsarin AI yana haifar da sabbin lahani da wuraren kai hari. Yayin da tsarin AI ke ƙaruwa, haɗarin hare-haren AI yana ƙaruwa. A cewar [14], kowane tabbatar da amincin wakili na AI, CVE amfani da rauni ana bin sawu, yana nuna buƙatar taka tsantsan yayin fuskantar barazanar da ke tasowa. Misali, hare-haren phishing da ke da karfin AI na iya zama nagartaccen tsari, ta yin amfani da algorithms na koyon injin don gujewa ganowa da yaudarar masu amfani wajen fallasa mahimman bayanai.
Haɗin kai tsakanin rashin lahani na kwana-kwana da ci gaban AI ya ta'allaka ne a cikin gaskiyar cewa tsarin AI ba su da kariya daga hare-haren kwanaki. A haƙiƙa, ana iya amfani da tsarin AI don ƙaddamar da nagartattun hare-hare waɗanda ke cin gajiyar rashin lahani na rana. AegisAI, kamfani ne wanda tsoffin shugabannin tsaro na Google suka kafa, ya samar da mafita mai karfin AI don dakatar da hare-haren mashin da AI ke tukawa [9]. Wannan yana haskaka wasan cat-da- linzamin kwamfuta tsakanin hare-haren AI da kariya. Don ƙarin misalta wannan batu, yi la'akari da batun harin fansa na NotPetya, wanda ya yi amfani da dabarun injiniyan zamantakewa na AI don yadawa da gujewa ganowa.
Don ba da fifiko ga tsaro ta fuskar rashin lahani na kwana-kwana da ci gaban AI, ƙwararrun tsaro dole ne su kasance da masaniya game da sabbin barazanar da abubuwan da ke faruwa. Wannan ya haɗa da hanyoyin sa ido kamar [4] da [6] don sabbin labarai da sabuntawa game da raunin rana-rana da ci gaban AI. Bugu da ƙari, dole ne ƙungiyoyin tsaro su ba da fifikon faci da sabunta tsarin, da kuma aiwatar da tsauraran matakan tsaro don hana hare-haren kwanaki. Misali, aiwatar da tsarin sarrafa faci mai ƙarfi na iya taimakawa rage haɗarin hare-hare na kwana-kwana, yayin da amfani da kayan aikin tsaro masu ƙarfi na AI na iya taimakawa ganowa da amsa barazanar da ke tasowa.
Za a tsara makomar tsaro ta hanyar tsaka-tsakin raunin rana-rana da ci gaban AI. Yayin da tsarin AI ke ƙara yaɗuwa, haɗarin hare-hare masu ƙarfi na AI zai haɓaka. Koyaya, AI kuma ana iya amfani dashi don haɓaka tsaro, kamar ta hanyar haɓaka hanyoyin samar da wutar lantarki ta AI don dakatar da hare-haren mashin AI. A cewar [3], Yuli 2026 ya ga ƙaddamar da AI guda 27 da aka rufe akan ThursdAI, wanda OpenAI, Anthropic, Google DeepMind, Meta AI ke jagoranta, a tsakanin wakilai, ƙirar iyaka, buɗe tushen, da coding. Wannan yana nuna saurin haɓakar haɓakawa a cikin sararin AI da kuma buƙatar masu sana'a na tsaro su ci gaba da gaba.
Sources
- https://techcrunch.com/2026/07/
- https://aitoolsrecap.com/Blog/AINewsJuly2026.aspx
- https://thursdai.news/releases/2026-07
- https://techcrunch.com/
- https://aiweekly.co/ai-news-today
- https://www.timesofai.com/news/ai/
- https://thirdruntime.com/
- https://www.secondtalent.com/news/tech/
- https://techcrunch.com/2026/
- https://nvd.nist.gov/vuln/detail/CVE-2026-58644
- https://www.cvefind.com/
- https://www.securityweek.com/new-check-point-zero-day-vulnerability-exploited-in-the-wild/
- https://www.cisa.gov/news-events/alerts/2026/07/22/cisa-adds-two-known-exploited-vulnerabilities-catalog
- https://axis-intelligence.com/ai-agent-security-incident-tracker/
- https://securityonline.info/