Never, because then the DEI group's budget would be cut. The incentives for the people actually running these programs are completely out of whack with what would be good for the company and for the people they're actually meant to help.

The qualitative objective for most companies should be something like: "Recruiting and hiring people with no bias against race, gender, religion, age, disability, etc... Treating those same people with no bias once hired, including pay, promotion, opportunities, and respect. Leveraging the diversity of perspective and skills of everyone in the company to maximize success of the company."

How do you measure that? If you're a SW company and you have 2% Black engineers is that good or expected? If its not good, how should you improve it?

I think these are legitimately important questions, but also exceptionally hard questions. I think the big problem though is that for the majority of the population there is little incentive to actually solve the problem. But I think money will eventually be what does it. Market inefficiencies will eventually lead people to want to solve this, but it can take a LONG time for these inefficiencies to manifest, since there are so many other factors at play. For example, look at college football. Alabama did not integrate black players until the 70s and they were fine until they played an integrated USC team -- and it took that long despite football being probably one of the places where inefficienes are squashed out pretty quickly.

I feel like this mindset is the same as CEOs reducing the IT budget because “We’ve recovered from our last critical outage and our systems are working fine now.”

I think there’s a valid place for a DEI-like group within HR ensuring a company’s hiring and promoting policies are fair in an ongoing manner.

There's nothing empty about that. It's measured, and evaluated.