Can you link to the paper you're thinking of? Maybe people are just talking past each other here. A biased random source can't bias the kernel random pool in any straightforward kind of way.
I explicitly said
> it becomes dangerous if they can preview the results or inspect the other sources
https://blog.cr.yp.to/20140205-entropy.html
I explicitly said
because the malicious source can just precompute the hash for the bias it wants.https://blog.cr.yp.to/20140205-entropy.html