In a sharded database, why is it typically a bad idea to rely on a standard database `AUTO_INCREMENT` column for your Primary Keys?

1. A Integer-based IDs are fundamentally slower for the database to search through than text strings.
2. B Managing a simple integer takes more CPU power than generating a complex UUID.
3. C Each shard generates its own sequence independently, which almost guarantees 'ID Collisions' (e.g., Shard A and Shard B both assigning ID `1001` to different records).
4. D Sharded database systems generally do not support any numerical data types for Primary Keys.