It probably depends on the jurisdiction, and if they interpret the smart contract as a computer program, and have hacking laws they can apply, or as an actual autonomous contract that is legally valid. Both seem like plausible interpretations, but I think the former is more natural/likely for the typical judge.