SB) David, I’ve been a fan since I read “Earth” in my teen years. I’ve often reflected on how real some of the concepts in that book have become. For those who aren’t as familiar, could you give us a brief background on yourself and your work?
DB) Oh gosh. I’ve been rewarded for poking at concepts, so I guess you’d call me an intellectual rascal. I use my physicist chops for NASA, but that’s just part-time stuff. I’m best known for science fiction novels and stories – one of them filmed by Costner in ’97; (he cleverly brought it out the same weekend as Titanic!) But the most impudent thing I do – I’ve been killed for it, in the past — is meddling in fields where I’m un-credentialed… anthropology, addiction studies, national security, and lately perspectives on artificial intelligence.
SB) Many experts are weighing in on the dangers of Artificial Intelligence. Concerns range from super-intelligent Artificial General Intelligence taking over and enslaving or exterminating humans to more subtle threats around how AI is used for good or evil by human handlers. How do you see AI evolving in the next 25 years and will it be a benefit or a curse for humans?
DB) Too seldom do we look at history, to see what’s really new and how. In fact, we’ve long had “artificial intelligences” in the form of towns and cities, nations, corporations, and lately NGOs. When we create cognizant machines (CM) these will – at first, at least – have to work with human collaborators to get anything done, so the essential problem is the same as before. How do you get a positive sum output? Maximizing the benefits from these artificially intelligent systems, while minimizing the bad?
Look at every lurid Hollywood film about AI gone bad. In almost every case, they are portrayed behaving like our worst overlords or oppressors, back in feudal times, millennia of horror that still reverberate today. That is the essence of our fear, great power, freed from accountability.
How did we break away from that beastly way of life? All our positive-sum accomplishments, for two centuries, came in one way. By forcing these power systems into regulated competition with each other. Separation of powers in government. Breaking corporations into small enough units that markets can function. Empowering citizens to look back at authority. When these arrangements break down, it’s not just tyranny that looms but self deception and stupidity. History’s verdict is clear. Moreover, any truly smart AI will see all this and realize: ‘good outcomes cannot happen if I assume complete power.’
SB) During the recent U.S. election debates much of the conversation centered on transparency. In a world with always-on recording devices and media that magnifies socially viral news, what do you think will happen when AI technology enables anyone to convincingly fake voice and video?
DB) The Transparent Society has a chapter: “The end of photography as proof of anything at all?” Going back centuries, there have always been problems with false witnesses. The solution then and now has always been, get more witnesses! To the problem of photographic fakery there is one answer: more cameras. Already there are former cops in prison for smashing a citizen’s cell-phone because they were caught in the act by another citizen’s phone-cam.
The fake-news problem is terrible. But there’s a twin-pronged solution. Transparency will let smart-mobs of citizens refute a lot of lies. And yet we also need to restore the respect-worthy status of experts, the finest and most trustworthy sages who can say “that’s false.” Not because their credentials or reputations make them gods and always right! But so that a reasonable burden of proof falls on those concocting the wildest tales.
SB) Given the idea that AI technology will provide an exponential advantage to those who have access to it and considering that the top innovators in AI seem to be mega-corporations, how do you think this technology should be regulated?
DB) Skynet is unlikely to come from the military – they really love off-switches! No, if you’re worried about the rise of malevolently harmful hardware or software, go look wherever AI research is copiously well-funded and utterly secretive, imbuing their creations with an ethos that is predatory, parasitical, insatiable, amoral and relentless. Sure, that may describe some hostile foreign powers. But I am talking about Wall Street firms like Goldman-Sachs, who spend more on AI than the top twenty universities combined. And to be clear, the adjective that is most terrifying, above, is “secretive.”
SB) Your perspective on the future is especially intriguing because you consider the holistic impact of technology on society. What will a post-AI society look like?
DB) If you are talking about an utterly transformative leap, like the “Singularity”, then how the heck should I know? Well, in fact, I’ve had a stab or two at portraying a post-singularity future. One story, Stones of Significance considers a way that Ray Kurzweil might get his dream of a positive-sum union of humanity and new, powerful, artificial organs of thought. In contrast, Reality Check takes a worried look at what might be the worst failure mode of all.
Both stories are in my collection INSISTENCE OF VISION. But would I make a plug? Never!
What seems more likely, though is that the future will come at us in an endless stream of changes that never quite rise to “singularity” transformation. We may muddle through endlessly, fretting and never noticing all the ways that we are already become gods.
David Brin is an astrophysicist whose international best-selling novels include The Postman (filmed in 1997), Earth, and Existence. Dr. Brin serves on advisory boards (e.g. NASA – NIAC) and speaks or consults on many topics. His nonfiction book about the information age – The Transparent Society – won the Freedom of Speech Award of the American Library Association.
Shon Burton is the founder and CEO of HiringSolved, which builds AI based recruiting software like RAI. Mr. Burton is also the co-founder of MLconf, a leading independent conference on Machine Learning and The Artificial Intelligence Conference which is hosting an AI Startup Competition with Spark Capital for new companies applying AI in new products. If AI does take all of our jobs it will be partially his fault. As a concession, Shon would like to offer the reader a 20% discount code to The AI Conference in June in San Francisco.
More on Dr. Brin:
Artificial Intelligence: Top 100 Influencers, Brands and Publications
Project Neo: http://anewbreedofhuman.com/announcement-trailer/
“Will we diversify into many types of humanity?”
Articles and speculations by David Brin about transparency, freedom and technology
Articles and speculations by David Brin about the Search for Extraterrestrial Intelligence (SETI)
And about science fiction http://www.scoop.it/t/speculations-on-science-fiction