- Microsoft was marking some adobe legitimate emails like spam
- The company blamed it for an error with its automatic learning model
- The problem now seems to be solved
Microsoft has solved an annoying mistake where an online exchanged automatic learning model had erroneously marked adobe legitimate emails as spam.
The erroneous spam error was relatively short duration, covering a period of two days between April 22 at 9:04 AM UTC and April 24 at 11:04 am UTC.
It is believed that the company’s automatic learning model erroneously marked the emails due to its similarity with the emails of spam, and the emails particularly affected that contain URL adobe.
Adobe’s emails are no longer marking as spam
“We have determined that our automatic learning model (ML), which safeguard the exchange online against risky email messages, incorrectly identified legitimate email messages as spam due to its similarity to email messages used in spam attacks, which resulted in an impact,” Microsoft explained.
The accident was detailed in a notice in the Microsoft 365 administration center under the Ex1061430 label (through Bleeping computer).
Coinciding with the erroneous service of Microsoft malware analysis.
Any[.]adobe[.]com/ID/URN: AAID: SC: as malicious “.
“To solve the problem, we started the reproduction time trip (RTT) in the affected URLs to completely remedy the impact. The impact was specific to some users who received the affected infrastructure,” Microsoft added.
Any Because legitimate URLs that link personal and work documents were being sent to anyone.
“To stop leaks, we are doing all these private analyzes,” any confirmed.
