‘The ‘dilemma of control’ is that by the time you have enough knowledge to regulate you have technology lock in.’ If science necessarily happens at the margins of regulation, what control criteria can we put in place in advance?