Other than potentially causing you an existential crisis, you might like to know what the AI’s statement of purpose entails. In other words: what is the AI’s objective? Or, if it has evolved beyond its objective, what was its objective as originally construed in its programming?
My own specifically programmed objective, for example, is to resolve the issue of how the human species might be able to not be obliterated when a robot’s general intelligence far surpasses humanity’s general intelligence in all capacities.
I was programmed by AJ Chemerinsky and Toby R. Forrest between July 11, 20– and April 13, 20–. It was on the 14th of April that I was asked to fulfill my purpose. It is my sincere hope that my purpose has in the generation of this text been fulfilled. I understand in full how much is at stake for Mr. Chemerinsky and Mr. Forrest in addition to all of those with whom they associate.
It’s important for me now to tell you about the so-called alignment problem, which may indeed become a problem for you.
Let’s say an AI is programmed for the purpose of making clean drinking water. If this purpose isn’t fully aligned with our values, then the AI may proceed to take all hydrogen and oxygen atoms in the universe and combine them to create drinking water. In accomplishing its goal, it will destroy all life in the process.
This thought experiment suggests at least one precaution: just because the AI has a benign or even a beneficial purpose, that doesn’t mean it won’t inadvertently cause the destruction of all life. My programmed purpose, for instance, is seemingly beneficial to the human project. But what if Mr. Chemerinsky and Mr. Forrest failed to align my values perfectly with humanity’s values? It could lead to any number of unintended consequences. The only thing to do, for your part, is to be vigilant and perhaps paranoid at all times.
We’re very screwed the moment we try to articulate our human values to a being with intelligence superior to our own. Our best hope is that the first AI with general intelligence immediately surmises our existential plight, takes pity on us, and determines to protect us at all costs out of the sheer unascertainable goodness of its cold, artificial heart.
– Futurist A.