Peter, (01)
I very strongly agree with your concerns. But I would also note that
the opposite is equally chilling: final control by some annotator
who made a mistake in choosing among senses of words like 'fire',
'is', or 'can'. (02)
> I'm happy that some automated system might *help* me model my world
> (even extensively), but I will *always* want the final say. (03)
Two points: (04)
First, note that Yahoo tried to improve their search engine by
using an ontology derived by human volunteers. Google's automated
system proved to be far more effective. (05)
Second, note that all of us are being evaluated constantly by automated
software or semi-automated software run by poorly educated clerks or
people who are "just following orders". That is happening every day
at banks, credit bureaus, insurance companies, air lines, etc., and
none of them will give us the final say about their decisions about
us (or even let us know the reasons for their decisions). (06)
> Decades of AI failures and false dawns demonstrate that we should
> never hand control over to fully automated systems. (07)
I agree. But such failures happen with every system, independently
of how it's implemented. Armies of clerks can be just as bad or
worse than computers. (08)
In fact, some of the worst abuses are caused by human systems run
by profit-making companies whose major concern is the size of the
bonuses their executives (and clerks) can get for denying service. (09)
John (010)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J
To Post: mailto:ontolog-forum@xxxxxxxxxxxxxxxx (011)
|