Artificial Intelligence researcher Hugo de Garis interviewed

[NLN]

Controversial Artificial Intelligence researcher Hugo de Garis tells us his thoughts about American science, "brain building," and superhuman intelligence in a new interview which can be read here.

[CraigD]

Hugo de Garis covers a lot of subject territory, both as a thinker and researcher and in this short interview! It’s hard to pick a starting place to discuss him or it.

 

Having read a lot of transhumanist/extropian fiction and non-fiction, I’m inclined to consider his “cosmicist/terran/cyborgian” speculation a minor footnote in this literature – which is not to say it’s less well though-out than that of folk like much less scientifically and mathematically literate and accomplished professional writer Charles Stross, but, IMHO, less readable and compelling, while being, in the final analysis, just as speculative.

 

All strong AI/extropy proponents – among whom I count myself – are confronted with profound obstacles to validating their intuitive beliefs, the slow progress – some might call it failure – of smart, talented, and well-funded researchers – including de Garis - in true, hard AI, and the Fermi paradox being high on that list. Perhaps for this reason, fiction remains the strong area of extropianism, while more “hard science” approaches remain at the cultural and scientific fringes.

 

Technically, I find de Garis’s work and commentary (of which I’d not been distinctly aware before this thread – thank’s, NLN :D ) fascinating. In particular, I’d like to know more about “reversible computing”, in which the information content of the actual circuits of a digital computer don’t destroy information, but increase steadily in entropy. This reminds me a lot of modern (1970+) resolution of the Maxwell’s demon paradox based on the Sizilard engine (described in a bit more length in this post). However, I’m unable (after a night of letting it sit at the back of my mind) to imagine how one would begin the actual design of such computing machinery. If any one better acquainted with the subject, or just plain cleverer than I, would sketch out a simple calculator or something along these lines, I’d be duly impressed. ;)

[NLN]

Craig, I've forwarded your intelligent comment on to Hugo. In the mean time, here is a link to a Wikipedia article about reversible computing.