Accelerating Future Transhumanism, AI, nanotech, the Singularity, and extinction risk.

4Nov/102

Katja Grace Honors Thesis Now Available

See the summary here, download it at the little box towards the lower right. Title: "Anthropic Reasoning in the Great Filter".

A major part of this effort is asking the questions, "what are different possible reference classes for anthropics/Doomsday Argument and what do they imply?", and "can we agree on updating our probabilities for being close to the Great Filter (whatever is responsible for the Fermi Paradox) if we aren't absolutely certain what reference class we're in?"

Read this first.

My current position is that it's extremely unlikely that life would develop to our stage because we live in a simple universe where even the evolution of consciousness is a miracle, but if it never happened, we'd never be around to observe it, so we happen to find ourselves in a universe where it did happen -- but just barely. Because there are many more simple universes (without life) than those with it (assuming whatever process generates universes in the multiverse generates more simple universes than complex ones), we should assume to find ourselves in one of the most abundant universes (we're typical, after all), we just happen to find ourselves in a universe that is common enough that it's simple, but complex (and consciousness-biased) enough that at least one conscious species evolved in it.

Is the universe really as simple as all that? Max Tegmark has published a paper arguing that the universe may in fact contain close to zero information.

Comments (2) Trackbacks (0)
  1. I’ve read the thesis in it’s entirety (I do odd things with my free time).

    Caitlin has written an outstanding undergraduate thesis but I think that her central idea that anthropic/indexicalreasoning can be used to predict the extinction of humanity is wrong.

    Her thesis is an expansion of Richard Gott’s idea which is as follows: We are among the first hundred billion (10^11)human beings to have existed. In the course of it’s lifetime, Humanity will expand and it’s population will exceed 10^20. There is only a one in a billion (10^(20-11)) chance for a random human to be born by 2010. How could we be so lucky? If, however, Humanity ends in the 21st century, the majority of all humans will have lived in the 20th or 21st centuries. We should assume that we are average, not one in a billion lucky, ergo Humanity will end soon.

    To see the flaw in this reasoning imagine 2 identical universes. U(1) is our universe where an interstellar asteroid crashes into the Earth in 2011 and destroys all life. U(2) is the exact same universe as U(1) except there is no asteroid and Humanity flourishes for billions of years and exceeds 10^20 population. In U(1) there is are versions of Gott, Bostrom and Grace who argue imminent extinction and they HAPPEN TO BE RIGHT. In U(2) there are versions of Gott, Bostrom and Grace who argue imminent extinction and they HAPPEN TO BE WRONG. Whether Humanity dies tomorrow or lives for eons, there will be people in the 21st century making identical anthropic arguments.

    Caitlin counter argument against the above point is found in section 2.3 of her thesis and consists of only 2 paragraphs. Her argument is that, if the above point is true, science would be impossible. This is clearly false. In U(1) and U(2), which are identical to each other and to our universe until the asteroid comes in 2011, people can clearly conduct scientific experiments.

    What exactly are “people” and “extinction” for antrhopic purposes? If the future contains numerous copies of the same person, or if we’ve partially merged minds with each other, how many people are there for Gott’s anthropic calculation? If we’ve boosted our intelligence by orders of magnitude do we still exist as “people” or have humans gone “extinct”? Reality does not recognize the categories that Gott relies upon for his calculations. Reality is nothing more than interacting particles and fields.

    Caitlin did, nonetheless, write a good and interesting thesis.

  2. Creating cool blog is not harder than creating great information. But you know that you done both incredibly well. Thank man.


Leave a comment

No trackbacks yet.