Microsoft Built a Prompt Injection Detector. Then It Caught a Phishing Campaign Instead.
11 Articles
11 Articles
Phishing campaign uses invisible Unicode to scale filter evasion
A newly tracked phishing campaign is sending millions of emails that embed invisible Unicode characters to bypass content filters and detection logic. The technique alters message text without visibly changing it for recipients, allowing...
Microsoft built a prompt injection detector. Then it caught a phishing campaign instead.
Microsoft flagged a phishing campaign last week that exploits a gap in how machines read text. Attackers are slipping invisible Unicode tag characters into email bodies; they don’t render on screen but change the underlying string that software processes. The company says attackers are already using the technique at scale to bypass spam filters and ML-based classifiers, and the same approach could cause problems for AI systems that regularly ing…
In a world where technology is increasingly used, cyber threats are also on the rise. One such threat is phishing campaigns that employ sophisticated techniques to bypass security systems. A phishing campaign sending millions of emails using invisible Unicode characters to circumvent filters has been reported by Microsoft. Microsoft Security Research stated, “Instead of using these characters to hide instructions from people while they are expos…
Mediafax
PSnewsVirtually invisible features for users are used in unsolicited message campaigns to bypass email filters, and Microsoft has identified a massive increase in the use of this method since February, according to a report submitted by the company.
For several months, hackers attacked US loan applicants with emails full of invisible Unicode characters gliding at the heart of keywords like "funding." For an antiphishing filter, these words no longer corresponded to those sought, remaining readable for a human. Up to 2.3 million messages were recorded in one day.
Coverage Details
Bias Distribution
- 50% of the sources lean Right
Factuality
To view factuality data please Upgrade to Premium
