If someone guesses what all the instructions are supposed to do and implements the wrong semantics, they didn't actually implement the same instruction set!
When all your security issues are violations of the spec, then it is not the language in the spec that is insecure.
You don't get to decide how an instruction set is used, and a 'low quality sandbox' is not a sandbox.
You've denied facts and continue to make both inexperienced and naieve claims that are dangerous. Not entertaining it further.