GAO report on the federal merit system — anyone read it?

There's a GAO report from earlier this year on "Strengthening the Federal Merit System in the AI Era" that's worth reading if you care about this stuff. Two things stood out: (1) they explicitly call out credential proxies as a known failure mode, and (2) they have no concrete recommendation for what to replace them with. I'd love to see the Risk Efficacy framework cited in the next round.

2 replies

I worked on the precursor to that report two years ago. The "no concrete recommendation" gap isn't accidental — they couldn't reach consensus on what good looks like. A vocabulary like Risk Efficacy's would have helped enormously.

The other gap in that report is that it treats "merit" as a single dimension. The whole point of the four-pillar framing is that you can't collapse it. Once you make that distinction explicit, the recommendation problem becomes much more tractable.

Sign in to join the community and add a reply.