The Dominance of Artificial General Intelligence – AGI The Final Chapter

Artificial General Intelligence – AGI might be the last invention of mankind. For better or for worse. Artificial intelligence is very beneficial to society in today’s age. AI will probably continue to be beneficial in the coming decades. However, as intelligent machines continue to improve, with the development of a true AGI or artificial general intelligence, there might be a serious threat to humanity. This issue is known as the AI control problem or the AI alignment problem. The risks involved in the process of improving intelligent systems might be intrinsic to intelligent machines that are goal-oriented. Today’s AI are known as narrow AI. They are able to beat humans in specifics tasks like chess but the same AI can not beat humans across multiple domains. At least not yet. A true artificial general intelligence might be able to solve some of the deepest secrets of nature. It might run simulations and come up with new mathematical equations and models to solve seemingly unsolvable problems in science. Such an intelligent machine might be able to cure any disease and create a wealth of sorts we’ve only seen in science fiction. It is difficult for many people to take the AI control problem seriously. Not just for the general public but also for researchers and some computer scientists who are themselves involved in the creation of tomorrow’s AI. The skeptics for this thesis do not bring forth any compelling argument why we shouldn’t be concerned about AI. To say AGI is far into the future and therefore we shouldn’t worry is a non sequitur. However long it takes to create true artificial general intelligence, if we do not destroy ourselves and continue to improve our machines, the day we will stand in the presence of an AGI will come sooner than we realize. Hopefully we will be ready.