Nick Bostrom talks in his book Superintelligence about the many dangers of AI. He considers it necessary that strong security mechanisms are put in place to ensure that a machine, once it gains general intelligence far beyond human capabilities, does not destroy humanity (most likely by accident). He describes this as a very delicate process that most likely will go wrong.
Considering that new technologies often neglect the necessary precautions and that this is highly relevant to national security, I wonder if there are already government agencies overseeing big technology companies like Deepmind. We are currently far away from an intelligence explosion or a technological singularity, but I would assume that governments want to have a foot in the door as soon as they realize and understand the dangers.
So my question is, what government agencies currently investigate and maybe even control AI development? The answer can be general or for a specific country if there is a big difference between countries.