The filter for inappropriate wording in German has been improved.
For some warning scenarios (e.g., unexpected recipients), AI Recipient Validation will no longer give out warnings in the first 2 weeks after a customer has been onboarded. This reduces the number of warnings in the beginning. This change only applies to newly onboarded customers.
Warnings about similar recipients are now only triggered in more relevant cases as the required match rate has been increased.
The filter for personally identifiable information has been improved. UK social security numbers are now identified more accurately.