Realistic [?] Routes to Substrate Independent Minds

Susan Schneider sent me this interesting article about a new group apparently devoted to unifying efforts to build artificial minds. 

Incidentally, the article contains a nice series of confused non sequiturs about computation and the brain:

“When it comes to the brain and the mind, the strong neuroscientific consensus is that behavior and experience, phenomena correlated with what we consider the mind, emerge from biophysical functions that are adequately described in terms of classical physics. These processes (and in fact, even quantum physical processes) are computable. It follows that the mind is computable; our brains are machines. The Church-Turing thesis implies that one Turing machine can implement another.”

(For an article that attempts to sort out at least some aspects of this kind of mess, see here.)

Update [8/27/10]: a comment by Joshua Stern, saying he is ok with the quoted statements, makes me realize that I need to be more explicit about the problems they raise:  (1) whether the mind emerges from biophysical functions that are adequately described in terms of classical physics is actually controversial (this is the least of my worries); (2) it is not clear what the authors mean by “classical physical processes (and even quantum physical processes) are computable”; (3) many disambiguations of the statement in (2) turn out false; the others turn out misleading or trivial; (4) whether something is computable is not the same as whether something is a machine; most relevantly, something may be an uncomputable machine; (5) to claim that the mind is computable because everything physical is computable (whatever that means, see (3), but clearly implying that this is the same as the computational theory of mind) trivializes the interesting empirical hypothesis that the mind has a computational explanation; (6) the final appeal to the Church-Turing thesis is irrelevant, except that in a statement that precedes the quote I give the authors define the Church-Turing thesis in a way that is at best misleading (leading to the fallacy of thinking that the Church-Turing thesis somehow entails that the mind is Turing-computable) and at worse false.  Oh, and the transition between the last two sentences sems to equivocally slide between “machine” and “Turing machine”.

For more discussions of some relevant issues, see also this, especially Sections 3 and 4.

One Comment

  1. Joshua Stern

    Well, I’m actually OK with that capsule, plus or minus some extensive qualifications.

    But for the project and the article, I saw the logo at the top and stopped reading. Allergy to this “singularity” talk since it escaped the Vernor Vinge novel.

Comments are closed.