About Rod D. Martin
Praise for Rod D. Martin
Book Rod to Speak
Politics & Economics
Technology & Markets
Faith & CultureEarlier today I posted a piece by Stephen Hawking on the dangers of artificial intelligence (more or less, it may be the last thing humans ever have a chance to create). Hawking frets over the sort of transcendent AI imagined by advocates like Ray Kurzweil. But Steven Omohundro suggests a different danger: that robots are by design anti-social and narrow-minded. In Omohundro’s view, AI poses the danger not of a runaway super-intelligence, but of a ruthlessly efficient, intelligent machine focused on a few tasks unrestrained by conscience or compassion, law and order, or even peer pressure. If this seems less dangerous, think again.
— RDM
Why There Will Be A Robot Uprising
by Patrick Tucker
Defense One
May 20, 2014
In the movie Transcendence, which opens in theaters on Friday, a sentient computer program embarks on a relentless quest for power, nearly destroying humanity in the process.
The film is science fiction but a computer scientist and entrepreneur Steven Omohundro says that “anti-social” artificial intelligence in the future is not only possible, but probable, unless we start designing AI systems very differently today.
Omohundro’s most recent recent paper, published in the Journal of Experimental & Theoretical Artificial Intelligence, lays out the case.
We think of artificial intelligence programs as somewhat humanlike. In fact, computer systems perceive the world through a narrow lens, the job they were designed to perform.
Microsoft Excel understands the world in terms of numbers entered into cells and rows; autonomous drone pilot systems perceive reality as a bunch calculations and actions that must be performed for the machine to stay in the air and to keep on target. Computer programs think of every decision in terms of how the outcome will help them do more of whatever they are supposed to do. It’s a cost vs. benefit calculation that happens all the time. Economists call it a utility function, but Omohundro says it’s not that different from the sort of math problem going in the human brain whenever we think about how to get more of what we want at the least amount of cost and risk.
For the most part, we want machines to operate exactly this way. The problem, by Omohundro’s logic, is that we can’t appreciate the obsessive devotion of a computer program to the thing it’s programed to do.
Put simply, robots are utility function junkies.
Even the smallest input that indicates that they’re performing their primary function better, faster, and at greater scale is enough to prompt them to keep doing more of that regardless of virtually every other consideration. That’s fine when you are talking about a simple program like Excel but becomes a problem when AI entities capable of rudimentary logic take over weapons, utilities or other dangerous or valuable assets.
In such situations, better performance will bring more resources and power to fulfill that primary function more fully, faster, and at greater scale. More importantly, these systems don’t worry about costs in terms of relationships, discomfort to others, etc., unless those costs present clear barriers to more primary function. This sort of computer behavior is anti-social, not fully logical, but not entirely illogical either.
Omohundro calls this approximate rationality and argues that it’s a faulty notion of design at the core of much contemporary AI development.
“We show that these systems are likely to behave in anti-social and harmful ways unless they are very carefully designed. Designers will be motivated to create systems that act approximately rationally and rational systems exhibit universal drives towards self-protection, resource acquisition, replication and efficiency. The current computing infrastructure would be vulnerable to unconstrained systems with these drives,” he writes.
The math that explains why that is Omohundro calls the formula for optimal rational decision making. It speaks to the way that any rational being will make decisions in order to maximize rewards and lowest possible cost. It looks like this:
In the above model, A is an action and S is a stimulus that results from that action. In the case of utility function, action and stimulus form a sort of feedback loop. Actions that produce stimuli consistent with fulfilling the program’s primary goal will result in more of that sort of behavior. That will include gaining more resources to do it.
For a sufficiently complex or empowered system, that decision-making would include not allowing itself to be turned off, take, for example, a robot with the primary goal of playing chess.
“When roboticists are asked by nervous onlookers about safety, a common answer is ‘We can always unplug it!’ But imagine this outcome from the chess robot’s point of view,” writes Omohundro. “A future in which it is unplugged is a future in which it cannot play or win any games of chess. This has very low utility and so expected utility maximisation will cause the creation of the instrumental subgoal of preventing itself from being unplugged. If the system believes the roboticist will persist in trying to unplug it, it will be motivated to develop the subgoal of permanently stopping the roboticist,” he writes.
In other words, the more logical the robot, the more likely it is to fight you to the death.
The problem of an artificial intelligence relentlessly pursuing its own goals to the obvious exclusion of every human consideration is sometimes called runaway AI.
The best solution, he says, is to slow down in our building and designing of AI systems, take a layered approach, similar to the way that ancient builders used wood scaffolds to support arches under construction and only remove the scaffold when the arch is complete.
That approach is not characteristic of the one we are taking today, putting more and more resources and responsibility under the control of increasingly autonomous systems. That’s especially true of the U.S. military, which is looking to deploy larger numbers of lethal autonomous systems, or L.A.Rs into more contested environments. Without better safeguards to prevent these sorts of systems from, one day, acting rationally, we are going to have an increasingly difficult time turning them off.
Rod is one of our nation’s leading minds. He’s smart, original, and very committed—a combination that one rarely finds.
Rod Martin is a Reaganite-Thatcherite conservative…[and] a man who has spent much of his life ministering to and working for the benefit of others. I know Rod as an advocate for freedom, a leader who understands the real changes we need in Washington, and who has the drive and ability to make those changes happen.
Rod Martin is the kind of dynamic leader who has the knowledge, ability and experience to lead us successfully into the 21st Century. His vision for America is compelling.
Rod Martin literally wrote the book on excellence. Humble yet brilliant, a man of love and compassion, with penetrating clarity he illuminates the world around us, and we are both stirred and enlightened.
Rod Martin is a born leader and an outstanding political mind. Anyone who takes him on is going to have a serious fight on their hands.
Rod Martin is a man of uncommon vision and character. Ever since I first worked alongside Rod in the early days of PayPal, I’ve been struck by his powerful vision for the future. Rod understands the intersection of technology, politics and culture in a way that few others do.
Rod Martin wears with uncommon grace an uncanny ability to link his vigorous faith to the practical challenges of both the business world and the political arena. Yet the respect he shows for others carries a warmth and modesty that are both too rare among those who are strong advocates of feisty positions.
America badly needs more leaders who really get the future. We need more Rod Martins.
Rod Martin is a modern day philosopher-statesman. He is a capitalist intellectual, a conservative iconoclast, a gentle adversary and a kind prophet. He inspires us towards a fresh national vision built upon the original founding principles.
Rod Martin is not only effective and efficient, but he is also perceptive, wise and gracious. He gets things done. He is a great friend and a great ally.
Rod is one of the sharpest, most gracious men I know, and one of the most gifted political thinkers today. His ability to analyze issues and communicate thoughtful insights on them have earned him the respect of people across the political spectrum.
I am proud to have been Rod Martin’s friend for a quarter century, and I can tell you that he is one of the finest men and most outstanding leaders in our church and in our country today. He has a knack for turning everything he touches to success.
Rod Martin is a cornucopian entrepreneur and eutropic thinker whose influence radiates through an array of companies, institutions, movements and churches. A renaissance American, he is also a prolific friend of freedom and faith around the globe.
If Rod Martin represents the next generation of leaders, America has nothing to worry about.
A leader and a servant, in public life and in private, Rod Martin strides at the top of his profession, and has nowhere to go but up. He is in the truest sense of the word ‘outstanding’.
Rod Martin is undoubtedly one of the most accomplished thought leaders today. He is a wonderful mentor, a humble role model, and a deep inspiration for the next generation of scientific innovators and entrepreneurs.
Rod’s achievements pale beside his character and his brilliance. He sees what others can’t. He leads when others won’t.
Rod Martin is one of America’s finest young minds.
The future of technology is the future of America. Rod Martin sees that future clearly, and also sees how to shape it for the betterment of all. He is firmly rooted in the traditions of faith and patriotism that have made America great. That’s why he has made, and will make, so many outstanding contributions in so many areas.
You must be logged in to post a comment.