Speaker Abstracts

Artificial Intelligence and Existential Risks

Nick Bostrom, Ph.D., Future of Humanity Institute, Oxford University

Close to 99.9% of all species that ever lived are now extinct. Will our own species suffer the fate of dinosaurs and dodos? How could that happen? When? And what can we do to stave off the end? An existential risk is one that threatens to annihilate Earth-originating intelligent life or permanently and drastically curtail its potential. Since we are still here, we know that no existential disaster has ever occurred, but, lacking experience of such disasters, we have not evolved mechanisms, biologically or culturally, for managing existential risks.

Attempts to quantify existential risk inevitably involve a large helping of subjective judgment. And, of course, there may be a publication bias in that those who believe that the potential risk is larger are more likely to publish books. Nevertheless, everybody who has seriously looked at the issue agrees that the risks are considerable. Even if the probability of extinction were merely 5%, or 1%, it would still be worth taking seriously in view of how much is at stake. In fact, it is sad that humanity as a whole has not invested even a few million dollars to improve its thinking about how it may best ensure its own survival. Some existential risks are difficult to study in a rigorous way, but we will not know what insights we might develop until we do the research.


Singularity or Dark Age? How the copyright wars threaten technological progress

Cory Doctorow, Science Fiction Author

Historical progress isn't inevitable - the pendulum of history doesn't have a regular period. Sometimes you get a 500-year Dark Age. The key prerequisite for technological progress is that our technology serves us. If our technology enslaves us instead, it will be an unparalleled, dystopian disaster.

Will our technology serve us, or control us? The answer isn't preordained: it depends on our conscious effort and eternal vigilance.


Productive Nanosystems: Toward a super-exponential threshold in physical technology

K. Eric Drexler, Ph.D., Nanorex

Nature demonstrates that productive nanosystems can work cleanly and inexpensively, converting common materials into billions of tons per year of intricate, atomically precise structures. Progress in molecular and nanoscale technologies has laid the groundwork for engineering simple productive nanosystems. These will enable the development of more intricate and complex productive systems, creating a feedback loop that drives accelerating change. Advanced productive nanosystems will deliver unprecedented products and productivity.


Trying to Muse Rationally about the Singularity Scenario

Douglas R. Hofstadter, Ph.D., Indiana University

In the past few years, a number of forward-looking thinkers have come to the conclusion that computer intelligence is rising so swiftly that quite soon, it will inevitably reach and then swiftly surpass human intelligence, and that at that monumental juncture in the history of this planet, humanity will have been eclipsed and replaced by its own creations. Such claims, put forth in a number of well-received books by such respected authorities as Ray Kurzweil and Hans Moravec, sound outlandish and possibly repulsive to many laypeople, as well as to many scientists and philosophers, while to others they seem very plausible, possibly even exciting and wonderful. In any case, this “singularity scenario” has been given spirited defenses in a growing number of books by serious thinkers.

Is it possible to gain any kind of 'objective' sense of the realism or unrealism of these possibly wild, possibly valid, perhaps wonderful, perhaps horrifying claims? The problem is that many disciplines are involved, including physics, evolutionary biology, neurology, psychology, computer science, philosophy of mind, and more. No one person has a sufficient mastery of all these fields to be able to see clearly into the future, not even a couple of decades, let alone a century or a millennium. This means that no one today knows the truth of such claims.

And yet there are some basic ideas that we should not lose track of, and that should help to keep us from confusing wild speculation with grounded reality. In my talk, I will attempt to chart out a way of looking at the “singularity scenario” with one's feet on the ground, and I will try to give, using my moderate familiarity with a number of different scientific disciplines, a personal appraisal of what I see as the likelihood of our being eclipsed by (or absorbed into) a vast computational network of superminds, in the course of the next few decades.


The Singularity: A hard or soft take off?

Ray Kurzweil, Kurzweil Technologies

Will the singularity be a soft or hard take off? My projections are considered radical by some observers because they are thinking linearly and leave out the historically correct exponential perspective. But in the singularity aware community, some thinkers consider my smooth doubly exponential progressions to be overly conservative compared to their idea of a "hard take off."


Being Good Enough

Bill McKibben, Author of Enough: Staying Human in an Engineered Age

The plausibility of a singularity gives rise to the question: why would we want one? While proponents and critic alike have grappled with the safety and environmental issues of new technologies, far fewer have bothered to consider whether it would represent progress or regress for human beings, simply assuming that More is Better. I will argue that human meaning is closely linked to our limitations – and indeed to our ability to limit ourselves – and that this meaning is precisely what technological change of this scale calls into question.


Cognitive and Emotional Singularities: Will super-intelligence come with super-wisdom?

Max More, Ph.D., Extropy Institute

Singularity scenarios typically revolve around the acceleration of computation and its effects on human and nonhuman cognition. We know that highly intelligent people can make terrible decisions. Their thinking may be driven by emotions of which they are only weakly aware or which they are unable to control or change. The higher the level of a person's cognitive intelligence, the greater their capacity for rationalizing and confabulating. Social influences can further distort our decisions and perceptions.

The question therefore arises: Will our emotional, social, psychological, ethical intelligence and self-awareness keep up with our cognitive abilities? How can we improve the odds of an affirmative answer turning out to be correct? Do the technologies driving the singularity inherently favor cognitive intelligence over emotional and social intelligence? What are the most promising means for furthering the evolution of our emotional responses and the wisdom of our decisions?

I will offer some thoughts on these questions, and suggest that we can do our part for a balanced singularity by adopting and adhering to evidence-based futurism and decision making (as embodied in the Proactionary Principle) and by encouraging others to do so.


Bringing Humanity and the Biosphere through the Singularity

Christine L. Peterson, Vice President of Public Policy, Foresight Nanotech Institute

With advanced nanotechnology and machine intelligence on the horizon, we face a future of vast change in our physical world and the world of the mind. But we need not abandon efforts to steer this future toward one which will work for both humans and the biosphere. We can identify ground conditions needed for success and reasons to hope that success is not implausible.

First, we should understand that the issue at hand is power: which entities hold it and what they do with it. Second, we can recognize that we already have models of how to handle entities more powerful than individuals. Finally, and surprisingly, we find that what we do today in computer science may determine our success in navigating the immense changes to come.


Searching for the Big Picture: Systems Theories of Accelerating Change

John Smart, Acceleration Studies Foundation

What systems models do we presently have for long-range processes of accelerating change? What general patterns can we observe for "phase change" singularities in physical systems? What is the paradigm of universal evolutionary development, and how might it help us understand and navigate an era where we find ourselves increasingly outcompeted by our technological extensions? Why is the systematic study of accelerating change so valuable, and presently so underfunded?

I will look briefly at accelerating change from universal, biological, human cultural, and technological perspectives. Some models of the past and speculations for the future will be proposed. I'll introduce a few well known and unorthodox ideas in acceleration mechanics, and some systems-compliant scenarios for a future of increasing intelligence amplification (IA) in human society, and increasing autonomous intelligence (AI) in our technology. I'll conclude with a very speculative prediction, and a call for greater scientific funding, transdisciplinarity, and acceleration awareness in an era of increasingly complex and powerful technologies.


Toward human-level intelligence in autonomous cars

Sebastian Thrun, Ph.D., Stanford Artificial Intelligence Lab

I will present results from a recent autonomous robot race, the DARPA Grand Challenge. This well popularized race required robots to drive without human assistance along a 131-mile long course through the Mojave desert. Our team from Stanford won this race. In developing the robot, our team developed significant new AI technology for robot perception and decision making. Our lab now focuses on making autonomous cars that can exceed beyond human driving ability in traffic. Since the Stanford Singularity Summit focuses on technology singularities, I will discuss where we stand in comparison to human level cognition in the driving task.


The Human Importance of the Intelligence Explosion

Eliezer Yudkowsky, Singularity Institute for Artificial Intelligence

In everyday life, we underrate the importance of intelligence because our social environment consists of only other humans, who as a species are far more intelligent than mice or lizards. The rise of human general intelligence enormously transformed the world. Yet we may have only begun to see the effects of intelligence. In 1965, the Bayesian statistician I. J. Good published a paper titled "Speculations Concerning the First Ultraintelligent Machine", in which he suggested that a sufficiently intelligent AI could redesign itself to make itself smarter, and then, being smarter, re-reinvent itself and become smarter still - a positive feedback cycle. Good labeled this the "intelligence explosion". An intelligence explosion could reshape the universe more than all human actions up to this point. It is the responsibility of this generation to shape the intelligence explosion.