攻擊者透過 Google 贊助搜尋結果推廣假「Plus 5.6」自訂 GPT,再把訪客帶到假 CAPTCHA 頁面,誘使他們執行 PowerShell 指令。
OpenAI 移除了一個惡意 GPT;但報道指,另一個變種兩日後仍然出現並維持運作。
How did attackers use paid Google ads and a fake “Plus 5.6” Custom GPT on the real ChatGPT site to lure Windows users through a Google SitesIllustrative image; it is not a screenshot of the reported campaign.
AI 提示
Create a landscape editorial hero image for this Studio Global article: How did attackers use paid Google ads and a fake “Plus 5.6” Custom GPT on the real ChatGPT site to lure Windows users through a Google Sites. Article summary: Attackers bought Google search ads that led to “Plus 5.6,” an attacker-created Custom GPT hosted on the real ChatGPT site. The real address made the lure look credible, but the GPT sent Windows users to a purported backu. Topic tags: general, general web, documentation. Style: premium digital editorial illustration, source-backed research mood, clean composition, high detail, modern web publication hero. Use reference image context only for broad subject, composition, and topical grounding; do not copy the exact image. Avoid: logos, brand marks, copyrighted characters, real person likenesses, fake screenshots, UI text, readable text, watermarks, charts with fak
openai.com
Google 搜尋 ChatGPT 時,部分用戶可能會在贊助搜尋結果中看到名為「Plus 5.6」的自訂 GPT。它開在真正的 ChatGPT 網站上,但並非官方產品。這個 GPT 會聲稱 ChatGPT 服務不穩定,叫訪客轉到一個所謂的備用網站;網站上的假 Cloudflare CAPTCHA 再誘使 Windows 用戶執行 PowerShell 指令,最終安裝遠端存取木馬。3617