How can I see why a "replace root volume" task is failing?

I've been trying to get a Windows EC2 instance back up that's been failing on boot, and I want to use the new AWS "replace root volume" feature that was released in 2021. I'm replacing the root volume from an AMI, but the "task" that gets created to replace the root volume keeps failing, and I don't know why.

Screenshot of the failed tasks: https://dl.dropboxusercontent.com/s/2h0tlizhsv9bmr6/chrome_q68K9S2Cox.png

In the docs, https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/replace-root.html, there are a bunch of requirements of the AMI in order for this to work. They are:

The AMI must have the same product code, billing information, architecture type, and virtualization type as that of the instance.

I'm sure that the AMIs I have used have the same architecture type (x86_64) and virtualization type (hvm). There shouldn't be a product code or billing information since I created these AMIs myself from machines that I built myself.

I just can't figure out why this is failing.