Comments

  1. static

    I’m a skeptic, and since the Summit is way too expensive for me to attend, I’ll post here instead.

    My main concern is whether we even have the hardware to simulate a human brain. An ordinary laptop or desktop obviously isn’t good enough, but what about the world’s fastest supercomputer? If you combine all the computing power in the world (desktops and laptops, playstation 3′s, embedded computing devices, etc) would that be enough hardware to simulate a human brain and potentially bring about the singularity? If not, how much longer would Moore’s Law have to continue so that the fastest supercomputer would match the human brain? Would it take until 2019? 2050? or 2100?

    The singularity idea seems too good to be true – if it were possible, there would already have been governments or large companies building artificial brains. The financial benefits of having them are huge.

  2. “My main concern is whether we even have the hardware to simulate a human brain.”

    At this point in time no, we don’t have the hardware to simulate a complete human brain.

    “what about the world’s fastest supercomputer”

    As of june 2009 that would be the DOE/NNSA/LANL Roadrunner, an IBM cluster that achieved 1.1 petaflops and could theoretically reach 1.4 petaflops. That’s 1.4 * 10^15 floating point operations per second. http://www.top500.org/system/9707

    “enough hardware to simulate a human brain and ”
    That’s debatable, and depends upon the level of detail at which you simulate: somewhere between 10^18 – 10^30.

    “how much longer would Moore’s Law have to continue so that the fastest supercomputer would match the human brain”

    By the end of the next decade supercomputers will exceed exaflop capability. I would guess it would be enough to see at least neuron level simulation of the entire brain by then. Blue Brain project actually has 2020 as timeframe I think.

    Ray Kurzweil’s famous prediction is that in 2029 we will have a computer that will pass the Turing test.

    “- if it were possible, there would already have been governments or large companies building artificial brains”

    There are many projects in fact but at this point it is mostly reasearch. So the job of universities (sponsored by companies), and R&D departments of large companies. I would name IBM as major player in the field.

    See http://www.philosophy.ox.ac.uk/__data/assets/pdf_file/0019/3853/brain-emulation-roadmap-report.pdf for much more information on this topic.

  3. I precisely had to say thanks once more. I am not sure the things I could possibly have created in the absence of those creative concepts contributed by you about that problem. Entirely was an absolute traumatic issue for me personally, nevertheless witnessing the very specialised style you treated the issue forced me to jump for gladness. I’m just thankful for the help as well as wish you comprehend what a powerful job you were doing instructing the others thru your blog post. Probably you’ve never encountered any of us.

Leave a Comment

Awesome! You've decided to leave a comment. Please keep in mind that comments are moderated.

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>