It is well established that infants show preferences for certain facial characteristics, including faces adults generally perceive as more attractive (source). I've also seen people cite studies on infant racial preferences as evidence that racism is innate.
I think there's an important distinction that's often being skipped over.
A preference is about putting something up.
Racism is about putting something down.
Those are not the same psychological process.
Preferring one thing doesn't inherently require devaluing another. I can prefer jazz over rock without thinking rock music is objectively worse. I can have a favorite color without believing every other color is bad. Likewise, an infant looking longer at one type of face than another doesn't automatically mean the infant is making a negative judgment about the other face.
Now, this doesn't mean early biases or preferences are irrelevant. They could absolutely interact with culture, upbringing, and experience as someone develops. But that's a very different claim from saying racism itself is innate. To argue that, you would need to show that these early preferences naturally become prejudice or discriminatory beliefs. Simply demonstrating that infants have preferences doesn't establish that connection.
So I don't think "babies prefer X" is convincing evidence that "babies are born racist."