@patrickcmiller

Like the case often presented by #AntiAi adherents, Stanciuc is cherry picking facts.

Firstly, she presents no supporting evidence for the headline "Ai....is making us worse at thinking"

There are two preprint studies, which antagonists use; the MIT EEG study with unpublished data values and tiny cohort, and the joke of a "study" that got a lot if traction because of a punchy title "cognitive offliading" garbage.
Notably, Stanciuc is smart by not using ANY evidence for her key headliner hook, because these two studies are the "strongest" support for her thesis, so she remains silent.

Second;
The #GoldmanSachs study is should be taken with a trailer full of salt. It's like a GS study on agriculture analysing human flesh sourced organic fertiliser. GS are some of the worst human beings on the planet, their every waking moment motivated by greed.

The "Ai has no macroeconomic impact" exactly mirrors the PC IT introduction to business.
There was no macroeconomic impact for 15 years.
So that's an argument, albeit not very strong by Stancic. Any analysis if "Ai has no macro impact" that does not mention Solow's Paradox (No Macro effect) is using incomplete, foundational data.

Third;
And this is the only credible argument she presents, there is evidence that Ai users suffer fatigue.
Its root cause is context switching.
And the reason for it us tools and workflows are not fully integrated with #Ai so the users have to switch to clay tablets and reed stylus.
Before I started using #Claudecode I was copy pastaing like savage which was super annoying.

"Using Ai is a learned skill"

#aislop #psychology #aieconomies #regulateai #llm