Security researchers have uncovered a large campaign of malicious Chrome extensions masquerading as AI tools, with more than 260,000 users installing them through the official Chrome Web Store. The campaign, known as AiFrame, includes at least 30 extensions designed to steal sensitive information while posing as legitimate AI assistants and productivity tools.
According to analysis from LayerX Security, the extensions use hidden server-side interfaces that function as privileged proxies. This design grants them broad permissions inside the browser, allowing attackers to scan active tabs, copy content, and intercept confidential data such as login credentials, financial details, and other personal information. The stolen data can then be transmitted directly to the operators behind the campaign.
Many of the extensions used names that mimic popular AI services, including variations of ChatGPT and Gemini, sometimes with intentional misspellings to avoid detection. Others appeared under generic labels such as AI Assistant or AI Translator. Among the most widely installed were “Gemini AI Sidebar,” “AI Assistant,” and “ChatGPT Translate,” each attracting tens of thousands of downloads.
Researchers say the attackers repeatedly reuploaded the extensions after removal by modifying names, IDs, and code. In some cases, previously banned extensions returned to the store within weeks, highlighting the difficulty of preventing persistent threats that exploit official distribution channels. Some of the malicious add-ons even appeared in recommended extension lists, suggesting that they successfully bypassed certain automated checks.
Security experts warn that relying solely on official app stores is no longer sufficient protection. Users are advised to scrutinize permissions requested by extensions, avoid tools that impersonate well-known services, and remove any suspicious add-ons immediately. Installing software only from trusted developers and using reputable antivirus tools can also help reduce the risk of data theft.