Reported by 2 outlets

Timeline

1. original reporting

   Hackers Abuse AI-Era ASCII Smuggling to Hide Phishing Content in Millions of Emails

   gbhackers.com · 39d ago · September 4th, 2026

   Threat actors have repurposed an AI prompt-injection technique known as ASCII smuggling to evade email security controls at massive scale, hiding invisible Unicode characters within financial phishing lures. Microsoft observed the activity reach more than 2.3 million messages per day, demonstrating how techniques first popularized in AI-security research can quickly migrate into conventional phishing operations. […] The post Hackers Abuse AI-Era ASCII Smuggling to Hide Phishing Content in Millions of Emails appeared first on GBHackers Security | #1 Globally Trusted Cyber Security News Platform .

2. update from cyberpress.org

   Microsoft Warns ASCII Smuggling Is Being Used to Hide Phishing Keywords From Email Filters

   cyberpress.org · 39d ago · September 4th, 2026

   Researchers observed a high-volume phishing campaign that inserted invisible Unicode tag characters into financial terms such as “funding,” “loan,” and “credit.” To recipients, the messages looked normal. However, the hidden characters split the words at the processing level, potentially preventing simple keyword-based filters from recognizing suspicious language. For example, the visible word “funding” could contain an invisible Unicode character between “fun” and “ding.” A user would still read the complete word, but a detection tool looking for the exact text string “funding” may fail to match it. ASCII smuggling is the use of non-rendering Unicode characters to conceal text inside apparently normal content. The technique commonly abuses the Unicode Tags block, ranging from U+E0000 to U+E007F. These characters are largely invisible in standard fonts and email applications. ASCII Smuggling Evades Filters The method became well known through AI prompt-injection research. Attackers can hide instructions in emails, web pages, or documents that humans cannot see but an AI assistant may process if it reads the raw content. Those hidden instructions could attempt to influence an AI system into exposing information or performing unauthorized actions. In this phishing campaign, however, the attackers did not appear to use the characters to conceal instructions for AI models . Instead, they used a single invisible tag character, including U+E0020, to break up financial lure words an