(This entry has been amended to correct mistaken interpretations of polling data.)
UPDATE: Boy, do I have egg on my face. Many thanks to alert reader Andrew Sisk, who pointed out that I might have been misinterpreting the data from our current poll. He’s right. The numbers I cited as indicating total agreement with a particular statement don’t really say that at all. The way the polling software works, the percentage actually reflects the relative number of respondents agreeing to each statement in comparison to other statements. So, although it’s somewhat interesting to see which futurist predictions have the highest (and lowest) relative acceptance rates, the poll doesn’t really tell us the percentage of our readers that do or do not agree with each statement. I apologize for the confusion and will work on finding a more reliable way to get the information we’re looking for.
I find it fascinating that so far only about one in twenty (6%, to be exact) a relatively small percentage of respondents to our current poll agree with the statement that “a technological singularity is certain to occur before the middle of this century.” Maybe that says something positive about IEET readers’ unwillingness to accept futurist visions that don’t pass critical review.
Our job at the Institute for Ethics and Emerging Technologies is not to champion a specific vision of the future, nor is it to predict that any particular amazing tech definitely will be achieved. Our role, rather, is to analyze and promote a set of principles that can help in guiding developments toward an outcome we favor.
It’s a pleasant surprise to discover that so many of our readers appear to share our skepticism toward unwarranted assumptions about the fantastic possibilities touted by a few of the more vocal transhumanists.
The poll is not over yet, but so far only one in ten respondents agree, for example, but at this point, the statement that “the first person to live a thousand years has already been born” is less widely accepted than we might have expected. And even fewer than that number are willing to say that “uploading of human personalities to computer substrates is bound to happen,” or that “some people currently preserved cryonically will be revived successfully,” or that “emerging technologies will produce a post-scarcity economy within 50 years.”
If you follow the chatter on some of the popular transhumanist blogs or mailing lists, you might think that everyone is convinced of all these things. But apparently that’s not the case.
Of course, this does not rule out the possibility that some of those things will be achieved someday. It just indicates what I would consider to be a healthy level of skepticism among IEET readers.
Also, as some of our commenters pointed out, many of the statements in the poll were deliberately phrased to emphasize the certainty that too often finds its way into these discussions. There is hardly any point in talking about what might come to pass if you’re not willing to consider outcomes different from what you prefer or what you expect.
In what should be an interesting and provocative event, IEET Senior Fellow Jamais Cascio will tackle this subject when he addresses the New York Futures Salon on Saturday, October 3rd:
With their unwavering focus on computing power and digital technology, leading Singularity proponents increasingly define the future in language devoid of politics and culture—thereby missing two of the factors most likely to shape the direction of any technology-driven intelligence explosion. Even if the final result is a “post-human” era, leaving out human elements when describing what leads up to a Singularity isn’t just mistaken, it’s potentially quite dangerous. It’s time to set aside algorithms and avatars, and talk about the truly important issues surrounding the possibility of a Singularity: political power, social responsibility, and the role of human agency.
He’s quite right.
Emerging technologies—whether AI or nanotech or synthetic biology—do not emerge into nor from a vacuum. They are developed within a context of political reality, amidst the daily tussle over regulation, funding, and proper usage. They do not arise fully-grown and pristine, but are hammered out, molded, shaped, and modified through endless discussions in corporate boardrooms and the halls of government.
Given that we live in a real world, not a science fiction world, where real governments and real companies make real decisions that affect real people—and knowing that we can’t say for sure when or if any spectacular new technology will turn everything upside down overnight—it is up to us to be engaged in current political debates and work out the best possible environments within which transformative technologies might emerge.
We at the IEET cannot actually see into the future and know for sure what will happen if certain choices are made. Nor, of course, can anyone else. The best we can do is to study the past, keep abreast of ongoing research activities and relevant current events, create models (or “scenarios”) of various possible outcomes, and then try to convince those in positions of power and influence to adopt policies that seem most likely to shift results in a positive direction.
Mike Treder is the Managing Director of the IEET, and former Executive Director of the non-profit Center for Responsible Nanotechnology.
Print •
Email •
permalink •
Discuss in Forums •
Send to: ¡
¡
COMMENTS
I think that you're right that IEET's comparative advantage lies in the area of values and policy, not prediction. On the other hand, one can't decide where to focus efforts, or estimate the expected value of policies, without at least an implicit probability distribution over the development and power of various technologies. Should we work on counterterrorism policy related to future micro-UAV designs or related to synthetic biology? The judgment will inevitably depend on estimated probabilities. If we don't know much, we should spread our probability mass widely to reflect our ignorance and rely on sources that know more, but one can't avoid the necessity of estimation.
"I find it fascinating that so far only about one in twenty (6%, to be exact) of respondents to our current poll agree with the statement that “a technological singularity is certain to occur before the middle of this century.” Maybe that says something positive about IEET readers’ unwillingness to accept futurist visions that don’t pass critical review."
Who is promulgating these "futurist visions that don't pass critical review"? What futurist would agree with that statement, or a less insane 90% confidence in such an event on that timescale? Not Kurzweil, Bostrom, Vinge, Moravec, Yudkowsky, or Hanson. Even someone absurdly confident about technological progress given continued survival of our civilization would still need to reckon with the possibility of catastrophic social collapse, nuclear war, biological weapons, or an asteroid strike. Some people may have clicked 'agree' to indicate that they think such a timeline more likely than not or something similar, but I don't know of any prominent futurists or 'vocal transhumanists' who would literally agree with that statement.
I think this would have been a far more interesting exercise if all the claims were ones that are actually seriously advanced. The claim about the first thousand-year-old is relatively good in this respect (although it might be more on target if one conditioned on the absence of human extinction or civilizational collapse). With respect to artificial intelligence, a better statement would be something like, "broadly superintelligent AI will be developed by 2066 with probability of at least X%" or an opportunity to enter a probability.
It is reassuring, though my first guess is that it's not really 6% of poll-takers that clicked that box but rather 6% of the total share of boxes clicked. Like for example, if all poll-takers unanimously clicked on all of the answers (which is an option in this particular poll), the poll would reflect that each answer had 10%, one in 10, since there are 10 answers each having an equal share of the 100%, (11 if you include 'none of the above'). So in that case 100% of people would have said they believe a technosigularity is right around the corner, but the poll results would say 10%. If something like that happened in this case then those poll-takers awaiting the nerd rapture could be way more than 6%. It's just that they favor the other stuff too. But then again, after typing this I am even more confused about it then when I started. So...
Please Carl, Probability/Prediction deals with only *one* aspect of intelligence...namely, rationalistic intelligence (ability to achieve goals). This views the world solely in terms of functional outputs. Rationality is not the be-all-and-end-all, and the irony is that those who *do* view it that way end up doing the very thing Mike warns of: uncritically supporting all manner of weird and extreme positions (just read 'Overcoming Bias').
There's another type of intelligence *creative intelligence*, based not on probability and utility, but on similarity and complexity. It's not about prediction, but about knowledge integration. It's based not on causal models, but on analogies. It isn't about optimizing goals, but about aesthetic principles. It's just as powerful (at least) as rational intelligence.
Despite what 'Less Wrongers' peach, real progress has always been a *combination* of these two types of intelligence: rationality AND creativity/passion working together, a rationalist-fundamentalist may preach strict-rationality in public, but I bet in private they secretly deploy the intuitive mode of analogical reasoning all the time to set their priors correctly And of course this 'priors setting' isn't Bayesian, not matter how fervently the rationalist-fundamental asserts it is. It relies on our values.
Seeing through extremist positions is not about predictions, its about principles. Mike's right.
Mike, I'm also interested in hearing where I could find notable people who believe "a technological singularity is certain to occur before the middle of this century.”
As an example of a singularitarian, I don't believe such things, and neither do such "more vocal" examples as Bostrom or Yudkowsky. Like me, I'm sure they'd consider deluded anyone who believed such a statement as presented in this poll.
In view of Andrew's comment, you corrected 6% with "a relatively small percentage", and similar things.
But it seems to me that with 11 questions, if all participants had answered yes to all questions, each question would have 100/11 = 9% of the total. So, perhaps the "relatively small percentage" is not small at all, is it? It could be very large for many or even all questions. Do you have access to individual replies?
In general, as in other cases, I mostly agree with the actual words in your post, but mostly disagree with (my interpretation, perhaps mistaken, of) their spirit.
Aleksei: I don't think my waking up tomorrow morning is a "certainty", as I realize that there is a certain probability (I hope small) that I may not wake up. How the hell could I claim that "a singularity is certain to occur before the middle of this century"?? Of course I answered no, which says nothing about my estimate of the likelihood, or the desirability, of a singularity.
This and other questions do not address real transhumanism, but the demonized caricature of transhumanism which some intellectually dishonest critics wish to sell to their readers.
Perhaps the confusions arise not from the poll and the statistics of the poll, but from the questions in the poll and both their subjective meaning and narrowness of margin for acceptance? Is it really relevant to argue and vote whether a technological singularity will happen this side of the mid century or the next? Or is it more relevant to find out if folks actually believe in it, and whether they sanction it or not? And even more important than this, is it better to discuss this or simply vote in a poll?
Aleksei is correct here to point out the finer detail concerning the questions in the poll, (and other polls).
One answer to solve the problems of your statistical representations as highlighted by Andrew, would be to add the selection of yes/no to each option in the multi-choice poll? This may give a more accurate picture of what folks actually believe and do not agree with?
@CygnusX1: Yes, I'm now working on a different kind of survey that will incorporate yes/no/unsure responses, which should be more useful in finding the percentages that we did not get in the current poll.
@Guilio: Is "real transhumanism" represented by statements like this one from Ray Kurzweil?
I and many other scientists now believe that in around 20 years we will have the means to reprogramme our bodies' stone-age software so we can halt, then reverse, ageing. Then nanotechnology will let us live for ever.
Because it is exactly that kind of over-the-top certainty that I am criticizing. Sorry if it seems like demonization to you; to me it's just being reasonably skeptical.
Kurzweil is not among the thinkers I'm a huge fan of, but I think you're reflecting what you *want* him to mean instead of what he actually means, when you claim that he is talking about certainties.
When people (especially scientists or technically inclined folks) say "I believe X", they often mean things like "X is more probable than the alternatives". They don't tend to think in terms of True Belief, even though you want to portray them as such.
So I also think you are indeed misrepresenting Kurzweil for your perceived political benefit, when you attempt to push the claim that he is talking about certainties.
Yes/no/unsure is still far inferior to letting people enter probabilities from 0 to 1, but would be an improvement.
I agree with Aleksei that it's outrageously uncharitable to interpret claims like that Kurzweil quote as meaning "X will surely happen with probability 1." The quote does indicate a probability much higher than I would assign, but Kurzweil is not ignorant of basic probability. If you've read his books, you should recall that he talks about various things that would prevent a near-term Singularity, e.g. WMD catastrophe, and makes arguments about their *likelihood,* while still assigning substantial probability to a non-Singularity outcome.
Criticism of Kurzweil shouldn't be about an imaginary straw-man of 100% confidence, which can be trivially refuted by an armchair philosopher or mathematician. Rather it should be aimed at Kurzweil's controversially high confidence (50%+) in his broad family of scenarios. The idea that there is a probability of 50% or more of a technological singularity within the next 25 years is a very strong (and minority) claim, but arguing for a lower probability requires actually engaging with the issues, drawing on evidence like the distribution of opinion of AI scientists, historical data on technology development and hype, etc.
@Mike: "Is "real transhumanism" represented by statements like this one from Ray Kurzweil? I and many other scientists now believe that in around 20 years we will have the means to reprogramme our bodies' stone-age software so we can halt, then reverse, ageing. Then nanotechnology will let us live for ever. Because it is exactly that kind of over-the-top certainty that I am criticizing. Sorry if it seems like demonization to you; to me it's just being reasonably skeptical.
While I am confident that indefinite life extension will be eventually achieved, I don't see it happening in 20 years. Perhaps in the second half of the century, and closer to the end. Perhaps even later. Very few transhumanists think practical, operational indefinite life extension will be a reality by 2029. Probably Kurzweil himself does not _really_ believe it: as Carl says, in his books there are qualifications if you look for them.
But Kurzweil's optimism is a refreshing change from today's often overly cautious, timid, boring and at times defeatist attitude. It reminds us that we live in a reality that can be reverse- and re- engineered if we push hard enough. It reminds us that our bodies and brains are not sacred cows but machines which can be improved by technology. He is the bard who tells us of the beautiful new world beyond the horizon, and dares us to go. This is how I choose to read Kurzweil and, in this sense, I think one Kurzweil is worth thousands of critics.
Our job at the Institute for Ethics and Emerging Technologies is not to champion (X) a specific vision of the future, nor is it to predict that any particular amazing tech definitely will be achieved. Our role, rather, is to analyze and promote a set of principles that can help in guiding developments toward an (Y) outcome we favor...and try to convince those in positions of power and influence to (Y) adopt policies that seem most likely to shift results in a positive direction.
If I was to say, for example, "Because of the looming Singularity, politics as usual will be obsolete in the not very long run," that would be taking a firm stance based on a specific vision of the future. And it wouldn't be very productive.
But if, instead, I said, "A society expecting to be transformed by powerful emerging technologies should focus on improving its science, engineering, and math education; should ensure government transparency and accountability; should promote citizen science policy participation; and should take a balanced approach to risks and benefits," that would be promoting principles deemed most likely to achieve a desired outcome.
IEET Blog |
email list |
newsletter |
The IEET is a 501(c)3 non-profit, tax-exempt organization registered in the State of Connecticut in the United States.
Contact: Executive Director, Dr. James J. Hughes,
Williams 229B, Trinity College, 300 Summit St., Hartford CT
06106 USA
Email: director @ ieet.org phone:
860-297-2376