There’s an article on Garmin’s website about the high-level makeup of the “running power” calculations: https://www.garmin.com/en-US/garmin-technology/running-science/running-dynamics/running-power/
It seem like it’s an imputed / calculated / implicit metric, unlike a bike power meter, ergometer, or even running foot pod power meter. So, it’s pretty similar to heart rate in that it’s a proxy metric – i.e., not a direct measurement of the amount of energy expended per unit of time. While there isn’t yet a great way to measure running power, there’s definitely a standard unit of measure: Watts!
The “potential power, power required to run up or down a hill” is the most compelling part of the Garmin Running Power, I think. Have you found that it’s been a comparable or better measure of your perceived exertion than heart rate stress + “fudge factors”?
The air resistance element seems hard to evaluate (does the local prevailing wind data really reflect the wind’s impact on your run?). And the “horizontal oscillation power” is an interesting measure of running efficiency. If you’re running inefficiently, you’ll have more oscillation and exert more power… but is that a helpful training impulse? Or just undesired extra stress?