Microsoft flags phishing spike using invisible Unicode smuggling

On February 26, a single phishing campaign pushed weekday volumes up to 2.37 million messages, each one peppered with an invisible Unicode character meant to break the words apart before a filter could read them, according to The Register.

Microsoft researchers Noam Kochavi and Sarah Wolstencroft published the findings on September 3. The technique is ASCII smuggling: "using invisible or non-rendering Unicode characters to hide content inside text that appears normal to humans," as Microsoft puts it. Instead of writing 'funding' in the email, the attackers wrote 'fun⟨U+E0020⟩ding'. That is a tag space wedged between letters, which a keyword filter chokes on but a human recipient never sees. It is the same trick that had picked up attention as a way to hide malicious prompts intended for AI models.

The pattern was distinctive. "After an intense first phase, with weekday volumes of 1 to 2.37 million messages, peaking on February 26, the numbers stepped down slowly to roughly 80 percent less per weekday by late March," the write-up says. "Most of these emails came from about 150 finance-themed sender domains, and they continued for the next three months." Weekends were silent.

"As AI-era attack methods become better understood, threat actors may adapt them for use in more traditional threats such as phishing and spam," the researchers wrote. That crossover shows up across our cybersecurity coverage, which has logged 253 stories in the last 90 days. Microsoft's fix is unglamorous: "Any content evaluated by keyword, signature, or regex logic should first have invisible and non-rendering Unicode code points stripped or folded, so that splicing them into a word no longer defeats the match."