Thanks for the suggestion, Luke. I hadn’t considered this. I’m going to have a go at the Open Phil Calibration Training applet first, and will scour the forum and Lesswrong for other useful training.
I’ve had mixed experiences using probabilistic language in my legal advice. It really depends on the client and the advisor being able to think like that. But I’ve got some internal clients who have responded well to this sort of advice—explaining things in terms of expected value can be especially good when giving advice about counterparties who won’t tell us what they’re thinking (e.g. in litigation or commercial negotiations). It would be excellent to work with EAs who think like this without prompting, and who actually expect that kind of advice.
Thanks for the suggestion, Luke. I hadn’t considered this. I’m going to have a go at the Open Phil Calibration Training applet first, and will scour the forum and Lesswrong for other useful training.
I’ve had mixed experiences using probabilistic language in my legal advice. It really depends on the client and the advisor being able to think like that. But I’ve got some internal clients who have responded well to this sort of advice—explaining things in terms of expected value can be especially good when giving advice about counterparties who won’t tell us what they’re thinking (e.g. in litigation or commercial negotiations). It would be excellent to work with EAs who think like this without prompting, and who actually expect that kind of advice.