8

Several prominent futurologists and technologists, most importantly Kurzweil and Moravec, have projected varying timelines for a forthcoming technological singularity in the next century or even few decades.

Assuming, perhaps naively, that we take them at their word, what might be some of the most significant, crucial or far-reaching philosophical implications of a massive increase in the quantity and quality of intelligence for our civilization?

In particular, what ethical considerations might be involved in the predicted development of simulated minds possessing great likeness to human minds -- or superhuman-level artificial intelligence?

  • 2
    Quick question: what sort of answer are you hoping for? Is it a speculative, "I think" answer (with reasons), or more of an "according to" answer (with citations)? – commando Mar 12 '12 at 19:29
  • 2
    @commando While I would certainly appreciate a robustly-cited answer, given the speculative hypothesis involved I am willing to settle for a well-reasoned response. In passing some of my major concerns are with the ethicality of reverse-engineering brain functionality and the civil rights of AIs (or uploaded mindfiles, etc.) – Joseph Weissman Mar 12 '12 at 19:46
  • 2
    I would think that the most significant point of singularity actually happening is the fact that it actually happened :) I find the prediction so confused and ill-founded that it would be a miracle if Kurzweil & co actually turned out to be right. That would need some philosophical explaining! – DBK Mar 13 '12 at 15:56
8

If they are right, the machines will answer this question far better than we can now. That will be the revolution in ethics.

The chance that this will happen on anything like the timescale Kurzweil and Moravec have suggested is so close to zero, however, that it's not really worth thinking about the problem. In a conflict between predictions based on sociology (the advancement of technological progress driven by market forces and incremental design improvements) and limitations based on physics (resistive dissipation of energy, leak currents, quantum mechanical uncertainty, etc.), they have bafflingly sided with sociology. This has pretty much never been true in the past. It's not that we don't know the physics, it's that we do know it and it won't support the kind of increase in processing power Kurzweil fondly imagines.

  • Thanks and +1! Could you expand a little bit on the physical limitations you think the Moravec/Kurzweil timelines are effectively setting aside? I have heard Michio Kaku for instance discussing the "collapse" of Moore's Law within a few decades -- I know it's somewhat beside the primary question, but I am curious whether this is the basic problem with the technological singularity hypothesis as you see it? – Joseph Weissman Mar 14 '12 at 22:03
  • 1
    @JosephWeissman - I don't have time to give an adequate answer, but there are all sorts of physical limits, ranging from the minimum energy required to perform a one-bit computation, through to maximum energy densities in all known materials, each of which will cause problems for various computing schemes. If you compare these things with e.g. the graph from The Age of Spiritual Machines (en.wikipedia.org/wiki/File:PPTExponentialGrowthof_Computing.jpg) it becomes apparent that Kurzweil is being a tad optimistic. Sigmoids look like exponentials early on...a classic blunder. – Rex Kerr Mar 18 '12 at 4:35
3

Regarding the ethics of simulated minds, it seems that The Turing Test cuts that particular Gordian Knot quite irrevocably. If the tenor of the debate concerning consciousness is anything to go by, a plausible alternatve has yet to be found to classify levels of consciousness, and the binary nature of the classification provided reduces all such ethical concerns to human equivalents*.

To expand on the first few lines of Rex's answer, though, I think the most exciting prospect is something analogous to Sam Harris's idea of a utilitarian calculus based on mental states. In a post singularity world his putative 'in principle' calculus becomes potentially an 'in practice' affair, which is a tantalising prospect in the extreme. In a world with the processing power and equipment to accurately quantify happiness- a necessary by-product of our ability to simulate minds- ethics becomes a different ball game all together.

*Certainly all such concerns I can imagine, but I think this might be worth asking as a separate question... Edit: So I've asked it

  • @TomBoardman - Interesting answer, and +1 (though I am not convinced of your main premise), but I don't think Harris has adequately addressed all the classic criticisms of utilitarianism (at least not publically), except to say, in effect, "Well, we can't get it right, but at least we can avoid doing anything massively and stupidly horrific." That leaves rather a lot to be desired IMO. If you know of more recent works of his that address some of the thornier issues (e.g. what exactly do you value, how do you weight individuals' competing concerns, etc.), please share! – Rex Kerr Mar 22 '12 at 9:03
2

In particular, what ethical considerations might be involved in the predicted development of simulated minds possessing great verisimilitude to human minds -- or superhuman-level artificial intelligence?

I imagine that such a development will split the world of ethics in two, each side with vastly different ideas:

On one side, people will claim that such simulated minds, by virtue of their verisimilitude to human minds, have civil rights. Physicalism, especially the more complete kind (everything is physical, i.e. a Materialist view) is likely to attract many to this view. Physicalists will see the simulated minds as equivalent to human minds; since they view all phenomena to be physical, they will think of a replication of the human mind with advanced AI to be equivalent to the human mind itself, possessing the same neurological connections and storage in the form of transistors and memory.

On the other side, people will claim that despite the verisimilitude of these simulated minds to human minds, they do not and should not have any ethical rights. This is because they are just physical objects, and lack the consciousness and sentience of human beings. Dualists in particular will be attracted to this view, because they do not view the human consciousness as replicable by purely physical means. Therefore, regardless of how powerful the AI of computers are, they simply will not have the minds of people, and so one cannot attribute any particular rights to them.

Ultimately, it seems to me that the ethics of computers and advanced AI is highly dependent on one's ontological views; those inclined toward metaphysics would be more likely to deny rights to computers, while the pure physicalists would be more likely to attribute rights to computers.

  • Belated thanks and +1! Some very helpful points here – Joseph Weissman Oct 3 '12 at 23:41
1

The main significant problem with the idea of the Singularity is that there is neither a good theory of consciousness and thus of intelligence; many of the advances in what might be broadly construed as artificial intelligence have been preceded by a good theory of what is possible and how it is been possible; for example:

  • indexing: this is the basic idea of search, and of course behind Google and other search engines; the vast increase of processing power and the availability of large data-sets ie the web tremendously enhances this capability; so much so that it sometimes feels like magic; but the basic idea is simple.

  • robotics: there is a good understanding of motion - Newtonian Mechanics; this can then be applied to the problem of imitating the natural motion of beetles, fishes, quadrapeds, bipeds and birds; a lot of this work has already been done; and various technologies can be seen to fit into a paradigm that will make this a reality.

  • sensory equipment: for example computer vision; we have a good theory of optics & acoustics; of chemistry and the electromagnetic spectrum; the significant new problem here is pattern matching - but again we can see quite directly what is required and its a matter of time that this will be done well.

Its also worth noting that the vast increase in processing power, which has been following Moores Law for some time is a consequence of the kind of technologies that are used within microprocessors; essentially everything becomes half the size every few years; and its this process thats been driving progress; a similar process for example cannot be envisaged for solar power simply because there is a theoretical limit on how much energy can be generated per square meter of sunshine; at some point Moores Law will stop and one will be forced into understanding Massive Paralellism; the web of course is a precursor to this at a very basic level; as are neural networks.

Before we even get close to human-like intelligence there will be significant ethical challenges associated with the convergence of certain technologies; for example drone warfare & mass surveillance have already caused much concern; similar problems will become apparent for the forseeable future.

Its also worth relating a story by Fredric Brown, named Answer; and short enough to quote here:

Dwan Ev ceremoniously soldered the final connection with gold. The eyes of a dozen television cameras watched him and the subether bore throughout the universe a dozen pictures of what he was doing. He straightened and nodded to Dwar Reyn, then moved to a position beside the switch that would complete the contact when he threw it. The switch that would connect, all at once, all of the monster computing machines of all the populated planets in the universe -- ninety-six billion planets -- into the supercircuit that would connect them all into one supercalculator, one cybernetics machine that would combine all the knowledge of all the galaxies. Dwar Reyn spoke briefly to the watching and listening trillions. Then after a moment's silence he said, "Now, Dwar Ev." Dwar Ev threw the switch. There was a mighty hum, the surge of power from ninety-six billion planets. Lights flashed and quieted along the miles-long panel.

Dwar Ev stepped back and drew a deep breath. "The honor of asking the first question is yours, Dwar Reyn."

"Thank you," said Dwar Reyn. "It shall be a question which no single cybernetics machine has been able to answer."

He turned to face the machine. "Is there a God?"

The mighty voice answered without hesitation, without the clicking of a single relay. "Yes, now there is a God." Sudden fear flashed on the face of Dwar Ev. He leaped to grab the switch. A bolt of lightning from the cloudless sky struck him down and fused the switch shut.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.