Copy Link
Add to Bookmark
Report
AIList Digest Volume 8 Issue 052
AIList Digest Tuesday, 16 Aug 1988 Volume 8 : Issue 52
Does AI Kill?
and
Free Will - How to dispose of naive science types
----------------------------------------------------------------------
Date: 9 Aug 88 14:05:30 GMT
From: Wayne Mesard <mesard@BBN.COM>
Subject: Re: does AI kill?
>From a previous article, by jlc@goanna.OZ.AU (Jacob L. Cybulski):
> The Iranian airbus disaster teaches us one thing about "AI Techniques",
> and this is that most of the AI companies forget that the end product
> of AI research is just a piece of computer software that needs to be
> treated like one, i.e. it needs to go through a standard software
> life-cycle and proper software engineering principles still apply to
> it no matter how much intelligence is burried in its intestines.
>
> I don't even mention the need to train the system users.
That's right, you don't. Neither do you mention the need to make the
specs and limitations of the system crystal clear to the customer as
well as to management and the people who will be touting the system and
its capabilities to potential users.
The Aegis was being expected to perform beyond specifications.
As another example:
The Challenger accident is almost universally recognized as a
politcally-motivated (in both the governmental, and organizational
sense) decision to employ a device in an environment for which it was
never designed. The shuttle's booster joints are not blamed, because
they were indeed performing to specification.
Why do I get the feeling that if the shuttle tragedy had resulted from a
software failure as opposed to a mechanical failure caused by the same
type of managerial errors, the public outcry would be against the
"homicidal software?"
Just because computer software possesses some rudimentary decision
making ability does not mean that it isn't bound by the limitations of
its human designers.
AI doesn't kill people. Politicians kill people.
> Jacob
--
unsigned *Wayne_Mesard(); MESARD@BBN.COM BBN, Cambridge, MA
You're living in a sick world when an album of cover songs by a
bunch of RAISINS can go platinum.
------------------------------
Date: 9 Aug 88 13:41:56 GMT
From: dswinney@afit-ab.arpa (David V. Swinney)
Subject: Re: How to dispose of naive science types (short)
In article <6657@well.UUCP> sierch@well.UUCP (Michael Sierchio) writes:
>
>Theories are not for proving!
>
>A theory is a model, a description, an attempt to preserve and describe
>phenomena -- science is not concerned with "proving" or "disproving"
>theories.
>A theory may or may not adequately describe the phenomena in question, in
>which case it is a "good" or "bad" theory
[deletion]
>Demonstration and experimentation show (to one degree or another) the value
>of a particular theory in a particular domain -- but PROOF? bah!
>--
I agree that theories are not for proving.
I would like to add, however, that a theory which does not adequately
describe a phenomenon should be called an "old theory" and not a "bad theory".
To be a theory in the first place, a hypothesis (at one time or another)
must have been shown to adequately describe a phenomenon or we would refer
to it as a discarded hypothesis.
It is becoming fairly common these days for scientists to present hypothetical
work to the public under the title "Theory".
This tends to dilute public trust in science and scientists in general.
We should all be a little more careful in our use of "theory" and the science
effort required by many parties to achieve that status.
------------------------------
Date: 12 Aug 88 12:37:26 GMT
From: ulysses!gamma!pyuxp!u1100s!castle@bloom-beacon.mit.edu
(Deborah Smit)
Subject: Re: How to dispose of naive science types (short)
In article <495@afit-ab.arpa>, dswinney@afit-ab.UUCP writes:
> In article <6657@well.UUCP> sierch@well.UUCP (Michael Sierchio) writes:
> >
> >Theories are not for proving!
> >
> >A theory is a model, a description, an attempt to preserve and describe
> >phenomena -- science is not concerned with "proving" or "disproving"
> >theories.
>
> It is becoming fairly common these days for scientists to present hypothetical
> work to the public under the title "Theory".
> This tends to dilute public trust in science and scientists in general.
> We should all be a little more careful in our use of "theory" and the science
> effort required by many parties to achieve that status.
Another big mistake is when scientists present hypothetical OR theoretical
work under the title "FACT". E.G. Evolution. The 'theories' of evolution
(of which there are many, many, and conflicting), do not even fit under
the title theory, since they are not demonstrable, and do not fit with
the facts shown by the fossil record (no intermediate forms -- before
you flame, examine current facts, fossils previously believed to be
intermediate have been debunked). It certainly cannot be called FACT,
though in college courses, some professors insist on speaking of
'the fact of evolution'. When evolutionists cannot support their
hypothesis by showing aggreement with known facts, they resort to
emotional mind-bashing (only foolish, gullible people don't believe
in evolution). Just my two cents. I enjoy reasonable theories,
they truly unify what we observe, but I don't appreciate emotional
outbursts on the part of those who can't give up their inaccurate
hypotheses to go on to something better.
- Deborah Smit
:
------------------------------
End of AIList Digest
********************