They're notoriously inaccurate, but I don't think that courts have given up on them yet. And if you think that's bad, just wait until you're given an inaccurate brain scan to tell whether you're lying.
https://www.sciencedirect.com/science/article/pii/S016926070...
>The rates of correct detection in guilty and innocent subjects were 86%, which was better than other previously used methods.
Eh, that sounds close enough to 'beyond a reasonable doubt', right?