> If you need protection against fault attacks in your threat model, you are way past the point where library safeguards will help you. This is probably the poi

> If you need protection against fault attacks in your threat model, you are way past the point where library safeguards will help you. This is probably the point at which you can't responsibly ship products without getting an actual crypto consultancy to sign off on it. I wish there was a way to make comment text bold, because this definitely seems like it needs to be emphasized.