Many red-teaming attacks don't rely on the identifiers at all, they are essentially "duck typing" the entities in the data model. Same approach is commonly used

Many red-teaming attacks don't rely on the identifiers at all, they are essentially "duck typing" the entities in the data model. Same approach is commonly used for dealing with data where a single identity may have multiple pseudonymized identifiers. There are a bunch of regulatory box-checking exercises like this that don't actually provide anonymization if the attacker is sophisticated.