Explain the theoretical concept of Recursive Self-Improvement as proposed by I.J. Good. Discuss the potential for an intelligence explosion where an AI becomes capable of improving its own architecture faster than human researchers can. Analyze the control problems associated with such a scenario.