[Fis] Howard AI chat, input to AI

Karl Javorszky karl.javorszky at gmail.com
Wed Aug 6 12:05:02 CEST 2025


Hi Howard,

Great engineering your work. Chapeau.

Now a question :
How does the Ai feed its basic knowledge? Have you fed it Lexika and
Handbooks?
If so, can you feed it new stuff, too?
Say you instruct your machine to open a new chapter, say contemporary
approaches to formal logic and natural philosophy.
There you could feed in
1. Perception of background / foreground contrast, a. Similar b. Diverse in
fg/bckg
2. Counting groups and sequences before one background commits an error
which makes the counting system in itself inexact/contradictory to the tune
of - 3x10-92 %.
2.1 show upper limits for sequences and for multidimensional partitions
2.2 discuss the =, <, =, >, =, <, << relation for collections with n : 1,
11, 32, 66, 97, 129, 137
3. Calculating next to the extent of any abstract symbol also the
certitudes about with which other abstract symbols it is in a common group
will add to insights about what keeps elements of groups together. The
concept is calculable if one uses the (ca 60.000) elementary connectors
known as cycles, on the etalon collection (which is 136 pairs of natural
numbers 1..16).

Maybe the idea will sound reasonable to a machine, even though the humans
teaching the machine abhor it like the devil a holy symbol.

Let us hope that the machines have no shy handling a new tool.

Best
Karl
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listas.unizar.es/pipermail/fis/attachments/20250806/ebe695c1/attachment.html>


More information about the Fis mailing list