This is the part of the test bias argument I cannot reconcile. I understand the claim that IQ tests were developed by people from a particular cultural background and may contain culturally loaded content. But I also know that predictive validity studies consistently show these tests predict academic and occupational outcomes at roughly the same rate across racial and ethnic groups. If the tests were truly biased against minorities you would expect them to underpredict minority performance. They do not. How do bias critics respond to this?
This is exactly the right question to ask and it is where the popular bias argument runs into serious empirical trouble 
Psychometric bias has a specific technical meaning that most popular discussions ignore entirely. A test is biased if it measures differently across groups or predicts outcomes less accurately for one group than another. By that definition the research is pretty consistent. Modern cognitive assessments show measurement equivalence across racial and ethnic groups and predict outcomes at roughly the same rate for different groups. What does exist are mean score differences between groups, but a mean difference is not evidence of bias. The test can be measuring accurately and still show group differences if the underlying developed abilities differ between groups for environmental or other reasons 
The bias claim usually conflates two completely different things 
A test that accurately measures real differences in developed abilities is not biased even if those differences are uncomfortable. Bias means the instrument is broken, not that the results are unwelcome 