Post #2405295
2026-05-11 12:04 UTC
@bodhipaksa@mastodon.scot
@bodhipaksa@mastodon.scot
I think Dawkins is not entirely wrong in saying he observes something that looks like intelligence. But I think the mistake is in the attribution. It is not the intelligence of the model he is observing, but rather the model is a representation of the intelligence of the people that has gone into training it - or should we maybe say: was distilled into a mathematical representation which can be used to compute certain problems.
Replies (1)
-
@rlcw@ecoevo.social 2026-05-11 12:04
That is fundamentally different from theists, who cannot refer to this well documented process. We know why genAI generates replies the way it does, because we know what the models were trained on and how their training was refined. It may not be public information, but it is available information. So I think Dawkings biggest sin here is mystifying what we are seeing, rather than acknowledging the actual process and the humanities collective labour stored in these models. @bodhipaksa@mastodon.scot