I can't tell whether you mean something much more trivial by 'dark arts' than I do, or whether you really think even PR on this small a scale is dangerously corrosive of our core epistemic standards. (Or is otherwise nefarious.)
As a toy example, suppose I don't usually wear suits, but I decide to wear a suit to a job interview or important meeting. Causally, this is useful in part because it's misleading; it signals more wealth, status, and professionalism than is entirely representative of who I am. But concluding that suit-wearing is 'dark arts' for that reason is rather melodramatic, given that it's so weakly and indirectly misleading, that it's perfectly normal for people to signal in this fashion, and that it isn't purely dishonest signaling. (Going to the trouble of publishing a book or wearing a suit does mean that you have more commitment and resources than a random person.)
Ditto 'going out of your way in any fashion to impress people on first dates is Dark Arts because it doesn't show them the Real You', 'putting extra money into making fancy business cards for your small business is Dark Arts because it makes you look like a bigger deal than you are', etc. There's a line we shouldn't cross, yes, but it concerns signaling that's much more extreme and unusual than any of these examples.
We're pleased to announce the release of "Smarter Than Us: The Rise of Machine Intelligence", commissioned by MIRI and written by Oxford University’s Stuart Armstrong, and available in EPUB, MOBI, PDF, and from the Amazon and Apple ebook stores.
Can we instruct AIs to steer the future as we desire? What goals should we program into them? It turns out this question is difficult to answer! Philosophers have tried for thousands of years to define an ideal world, but there remains no consensus. The prospect of goal-driven, smarter-than-human AI gives moral philosophy a new urgency. The future could be filled with joy, art, compassion, and beings living worthwhile and wonderful lives—but only if we’re able to precisely define what a “good” world is, and skilled enough to describe it perfectly to a computer program.
AIs, like computers, will do what we say—which is not necessarily what we mean. Such precision requires encoding the entire system of human values for an AI: explaining them to a mind that is alien to us, defining every ambiguous term, clarifying every edge case. Moreover, our values are fragile: in some cases, if we mis-define a single piece of the puzzle—say, consciousness—we end up with roughly 0% of the value we intended to reap, instead of 99% of the value.
Though an understanding of the problem is only beginning to spread, researchers from fields ranging from philosophy to computer science to economics are working together to conceive and test solutions. Are we up to the challenge?
Special thanks to all those at the FHI, MIRI and Less Wrong who helped with this work, and those who voted on the name!