UC Berkeley professor Stuart Russell warns that the unchecked development of artificial general intelligence could pose a serious existential risk, including a significant chance of human extinction.