DeepMind believes reinforcement learning is enough to develop a strong form of artificial intelligence