From Newsgroup: talk.origins

https://abcnews.go.com/GMA/Style/controversy-stirs-ai-generated-models-new-guess-ads/story?id=124271323

The fashion magazine Vogue is taking flack for publishing an ad using an
AI edited image. The fashion model in the ad doesn't really exist, but
AI editing likely means that there was a fashion model or models used to produce the original digital image or images, and that AI likely
manipulated that to produce the final image.

AI essentially manipulates the existing pixels (digital position and
content) to modify the image. It takes the basic form and then changes
the 2D image. In order to do this it has to change the information of
each pixel involved.

Variant lifeforms do not evolve in this way. What the organism looks
like is dependent on the interaction of the animals genetics with the environment, and that environment includes the interaction of cells in
the embryos of multicellular plants and animals. Phenotype can be
physically manipulated as AI does an image. Plastic surgery, some
cultures have manipulated the shape of the skulls of their children, and
you can do things like mess with development by placing a sliver of mica between cell layers of a frog embryo. Some of the environmental effects
can be inherited by the next generation by the reversible modification
of the DNA or chromatin, but since the sequence of the DNA has not been changed the inheritance is fleeting, and Lamarkism is still dead. The imprinted effects usually do not reproduce the environmentally induced pheontypes that caused the imprinted changes. The human example has
been the Dutch starvation during WWII. Children were stunted and did
not develop properly. The next generation was not stunted, but suffered metabolic issues and health issues. It seemed that their genes were not
as well adapted to a decent diet.

The DNA has to be altered in order to make lasting changes to the
phenotype of the organism. Any change has to work within the system
that is already working. After any change is made the environment gets
a say in whether or not the changes are acceptable. If the change is deleterious to the survival of the organism in the environment that it
exists in it gets selected against. If the change doesn't change the organisms ability to survive in the existing environment very much it
can drift in the population. If the change has some advantage it can be selected for and may quickly take over the interbreeding population of organisms. Life has not evolved the way that AI images are created.
That is the issue that the ID creationist scam has faced since day one.

ID perps like Behe have understood this from the start of the ID scam.
Behe's IC three neutral mutations occurring in a system to produce a new function depends on how biological evolution has had to accomplish
change for billion of years. You have an existing system, most
mutations are neutral, some are deleterious, and a few might do
something interesting. This is just how nature has worked for billions
of years. Behe claims that 3 specific neutral mutations are so highly unlikely to occur within a limited period of time that they are
basically impossible. This is unfortunately not true. If something can happen and it does happen, there is no do over, it becomes something
that just happened. Not only that, but Behe has to accept that this is
just the way things occur in nature because he has found examples of 2
neutral mutations occurring within his time limit to produce a new
function, and he knows that such things would be routine occurrences in organisms with a population of, at least, 100 million, and his examples
have occurred in populations of trillions. Behe has never looked for
examples in his IC systems, likely because he doesn't want to know the
answer that he expects to get. He has never tried to verify if his IC
systems are his type of IC. So far, intelligence hasn't been needed to
evolve the systems that Behe knows evolved on this planet.

Ron Okimoto

--- Synchronet 3.21a-Linux NewsLink 1.2