I've recently looked this stuff up, but I don't have the energy to provide all the citations right now. Another recent answer of mine has most of them. I'm just going to do the black vs white comparison--each group has its own intricacies to consider, and I've only considered this one.

In academia, the disparity in numbers of black full professors compared to the demographics overall is about 1:3 underrepresented (4% professors, 12% population).

In academia, the number of professors--full, tenured professors--is roughly at parity with the number of black Ph.D. holders at the relevant point in history. So presently, the faculty level has pretty much caught up, consistent with what you'd expect from no more than minor discrimination since the Civil Rights Act. This is not the source of the disparity.

Ph.D students of all races finish at similar rates these days. Black students are slightly lower but it's a couple percentage points off the average--nothing alarming. This explains very little disparity.

It's very hard to find solid numbers on PhD applications because there's no uniform application process--we don't know how many people apply to how many universities, so we might count the same student many times. I see lots of anecdotal evidence that grad schools are desperate for such students, so if there is a disparity it's probably not the fault of the hiring committees. But it could be that black students are either difficult to persuade to enter a PhD program or have been discouraged from entering a PhD program (or both).

But how do you even get to the point of entering a PhD program?

The 4-year college graduation rates are significantly different, with blacks about 20% lower (60-40 overall). Furthermore, among those who do graduate, the GPA for white individuals is substantially larger than for black individuals (3.2 vs 3.0)--for the top tiers who are likely to progress to more advanced degrees the disparity is even larger (2x for 3.5, which is often a bare minimum for grad school). Another metric is the average GRE scores (https://www.prepscholar.com/gre/blog/average-gre-scores/) which are about 2/3 of a standard deviation lower for black people than white--and you only put yourself through the GRE if you're seriously thinking about grad school. The math and English disparities are very similar, suggesting that biased content is unlikely to be the cause; rather, there's a disparity in ability.

Okay, so why the low graduation rates, and why the apparent lack of talent at that point? Well, whites have a somewhat higher rate of enrolling in college (62% vs 56% in 2019)...but not nearly as large as the SAT score disparity (5x for 650+ on math, 3.5x for 650+ on verbal--because of the small numbers of questions that distinguish the top scores, I'm generally uncomfortable reasoning about scores over 650). Note again that the math disparity is worse, which is opposite of what you'd expect if the test was intrinsically biased. Also, the mean scores are off by...about 2/3 of a standard deviation, just like the GRE. So you have unequal abilities going in, and therefore it's not quite such a shock that it's not equal going out.

So far, academically, we're basically tracking a difference in ability all the way through from graduating high school, with a possible additional problem with motivating people to apply to grad school.

Now I'm going to cut to the chase because we have a shortcut instead of following everything down to parental income, parental involvement in schools, breastfeeding, reading to children, presence of fathers in homes, and so on, and all the factors that after correction indicate that race isn't correlated with IQ to any substantial degree. Eventually you get to the answer of "it's not race, intrinsically". But you have a heck of a lot of small messy factors on the way, very few of which have anything to do with stealing from people now.

The Harlem Children's Zone is absolutely amazing. They take children (mostly black) in some of the poorest areas and erase or even reverse any racial disparities. They have a preposterous 97% college acceptance rate--if not going to college is a failure that's ten times better than the white average.

But it's resource-intensive.

That's the predominant difference specifically for academia. It's investment. It's not stealing from who is already here, it's missing the investment in future generations. Serious, thoughtful, loving, effective investment. Yes, the patterns of lack of investment were set up by past thefts and discrimination, but at least in academia that is mostly a thing of the past in terms of outcomes (not in terms of personal experience--there are still enough racists to make things unpleasant, but not, apparently, by the numbers, to hold people back particularly much). You can see the same phenomenon in the drastic academic advantage (yes, that's right, advantage!) home-schooled children have over children taught in either public or private school--and this is even with the religious nutjobs averaged in. Serious, thoughtful, loving, effective investment. That's what it takes.

Now, the question is: if the disparity is in investment, how do we fix it? We can't expect impoverished or otherwise stressed people to do it on their own--if we're holding them back at all we need to quit it, but that alone likely isn't enough. We need external investment too. Do we invest by race, or do we invest by need? And who funds it?

My tentative answer is: invest by need, and naturally the racial disparities will be targeted; and those with accumulated wealth should fund it (they are most able to pay for a long-term investment), and naturally the racial disparities there will be evened out too.

Fair in application, fair in funding, and has a track record of working (HCZ), unlike screaming at people that they're stealing everything even including individuality.

(The answer for Fortune 500 companies is different and more damning. Also, my answer only covers being in the ballpark of the level of disparity. There may be an additional impact of prejudice--there probably is, since there usually is when you look for such things--but it is not the dominant factor in academia overall. It might be the dominant factor in some disciplines, however.)