Malware via ChatGPT: Fake Outage Pages Exploit OpenAI

Threat actors are exploiting ChatGPT's content-sharing infrastructure to spread malware. This new campaign, called "LLMShare," tricks users into downloading malicious software. Here's how it works: Malicious ads on Google Search target users looking for "ChatGPT download" or "OpenAI desktop app." When clicked, these ads redirect users to fake outage pages hosted directly on legitimate chatgpt.com URLs. These pages tell users the website is down due to high traffic and urge them to download a desktop app. Because these fake pages are on OpenAI's own domain, they look legitimate, bypassing common security warnings. This makes it much harder for users to identify the scam. This discovery shows that generative AI platforms are becoming attractive tools for cybercriminals to bypass security. It highlights the growing need for vigilance even when links appear trustworthy.