From: Jim Fehlinger (fehlinger@home.com)
Date: Wed Jan 17 2001 - 06:24:55 MST
"Eliezer S. Yudkowsky" wrote:
>
>
> I've changed positions quite a bit over the years...
>
> (Some positions that have altered include:
> ...
>
> 3) When the Singularity will occur - moved from 2025 to 2008.)
Hooray! Glad to hear the nay-sayers haven't worn you down.
Before encountering your Low Beyond articles in '97, I had
already:
1. Encountered the usual million-year timescales for the
replacement of humans by their successors in authors such as
Stapledon (_Last and First Men_, _Star Maker_) and Clarke
(_The City and The Stars_).
2. Encountered near-future transcendence scenarios whose
plausibility as anything but entertainment vehicles was
weakened by reliance on extraterrestrial intervention
(Clarke's _Childhood's End_ and _2001_).
3. Found a friend, whose intelligence I respect a great
deal, who took it for granted that within a thousand years
humans would have been replaced by some sort of amalgam of
cybernetics and genetically-engineered "stuff". His name
is Joe Fineman (he has posted to this list a few times in
the past, but not recently), and he's the **only** other
human being I've encountered in the flesh who actually
thinks about these things. Unfortunately, his thousand-year
time-frame makes it impossible for him to take seriously,
or even suspend disbelief enough to read for pleasure,
**any** science fiction that's set in the far future but has
human beings in it (such as Clarke's _The City and the Stars_).
4. Read Vinge's _Marooned in Realtime_ and first heard
of the idea of the Singularity -- a sort of runaway inflation
of technological progress. (I also read _A Fire Upon the
Deep_, was amused by the extrapolation of an interstellar
Usenet, and became familiar with the terminology therein --
Low Beyond, Transcend, Powers, etc.).
5. Read Moravec's _Mind Children_, and delighted in the
spectacle of an apparently reputable scientist being allowed
by a reputable publisher to set forth a timetable for the
creation of artifacts of human-level or greater intelligence
within the next 50 years, in a serious work of non-fiction.
However, it was Yudkowsky's Web articles in which I first
encountered an apparently sane, intelligent, and articulate
person writing with panache and plausibility (and with a distinct
disinclination to pull punches for the sake of sparing the
reader's feelings, perhaps even a positive delight in
future shock "pour epater la bourgeoisie") that all this
might (nay will, by Vinge) happen within reach of my own lifetime
(with a bit of luck -- on my part, that is). Heady stuff, and
it sure got my adrenalin flowing when I encountered it, though
my acceptance of its plausibility waxes and wanes with the
moon. I was instantly recruited as an agent of memetic replication
for this work, though my success in turning on other folks to
singularity.posthuman.com has been rather pale, even among
fellow computer programmers (very few even of computer programmers,
it turns out, are true technophiles or potential Singularitarians,
at least in the circles I find myself in at present -- this has
been a big disappointment to me!).
First it was to be 2035 (IIRC). That would be a stretch for
me (though I'm in reasonably good shape, apart from a touch of
arthritis) -- I'd be 83 that year. Then 2025. A more
comfortable target, for this baby boomer. Now 2008, you say --
before I've even retired or paid off my mortgage!.
2008 is near enough that I can almost taste it. Let's see,
what do **I** expect for 2008?
1. On the computer front, we'll be in the 2005-2010 timeframe
for those next-generation chip-manufacturing technologies that
Intel and IBM were touting recently -- extreme ultraviolet
(EUV) lithography, .07 micron line widths (IIRC), 10 gigahertz
clock rates, etc. Perhaps 64-bit processors will be in the
mainstream by then, if the Itanium doesn't flop. This will all
be on the typical desktop, along with RAM in the gigabyte
range, hard-disk storage in the hundreds of gigabytes. With that
kind of processing power on the desktop, decent multimedia even
via low-bandwidth Internet is practical.
2. On the networking front, we may have really-high
bandwidth, guaranteed-service Internet 2 for the national
labs and big universities -- something that could support
surgery by telepresence, for example. Probably no more people will
have access to this in 2008 than had access to the Internet in the
early 80's. On the consumer front, Internet adoption
will probably have reached saturation (in the U.S. at least), at whatever
percentage that turns out to be; broadband penetration will still be
creeping along, and decent-speed, relatively cheap wireless Internet
access may prove to be the next big consumer app. I guess my
office (at work) will be wired with gigabit Ethernet.
3. On the bio front, I have no idea. Anti-aging therapy?
Smart drugs? Better treatments for cancer, heart disease,
Alzheimers? Who the hell knows? But my expectations are rather
conservative on this front.
4. I don't expect SETI@home to contact any aliens, in the Virgo
Cluster or anywhere else :->
Those are my surprise-free projections.
Now lets see, Singularity, as described by Yudkowsky (IIRC), means
a runaway positive feedback loop engendered either by machines
having become sufficiently intelligent to design their own more-intelligent
successors, or IQ-boosted humans designing even more powerful IQ
boosts for **their** successors, or a combination of both -- skating
past slippery questions such as the meaning of "intelligence", whether
computers have any meaningful similarities to brains, etc.
Can I see a Singularity developing from my own surprise-free expectations
of 2008? Well, it would have to depend on moderately better-than-present
hardware combined with moderately better-than-present networking combined
perhaps with a software breakthrough of some sort that would result in the birth of
genuine AI.
In my very humble opinion (I am not Paul Saffo, Jaron Lanier, or anybody
else ever likely to be interviewed by John Dvorak on Tech TV's "Big Thinkers")
I just don't see it happening. I think the technology of 2008 will still
be too crude and macro-scale to compare with the fine-grained complexity
of biological machinery exhibiting the full range of behavior to which the
adjective "intelligent" (whatever the hell it means) is usually applied. This would
still be true even if you could aggregate all the computers on the planet,
and will, I think, remain true in 2008.
Now, that's the surprise-free scenario. There is another possibility:
5. There will be a major paradigm shift of some sort. A breakthrough
into the third dimension of micro-scale electronic circuitry, as Kurzweil
anticipates. Molecular-scale computing. Optical computing. Quantum
computing.
This is the sort of thing on which I focus my expectations of Singularity.
"Cherchez la hardware" is my motto (please forgive my French).
Perhaps Eliezer would like to give us a clue as to which basis he's
counting on for the 2008 runaway -- (1) a software breakthrough on
somewhat-better-than-today hardware and networking, or (2) a major
paradigm shift in the hardware. Or something else? My guess for Eliezer is (1),
because of his writings on seed AIs, but my hope is (2) -- though even
if strong indications of such a shift have appeared by 2008, I wouldn't
expect to have reliable and cost-effective manufacturing processes or
commercial products by then.
However, I'd love to be surprised!
Jim F.
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 08:04:57 MST