How Algorithms Produce Social Intelligence site 
YOUTUBE L5ptc9WWcJc Why Algorithms Can’t Think: “Artificial Communication” by Elena Esposito
> 20:03
... predictions produced by algorithm are with the shape they have the form they have and it's also the form of the aggregate prediction is very similar to the form of divinatory prediction uh some features for example our forms of prediction in our modern culture are always probabilistic …
20:59
although i don't tell you anything about averages they tell you very potential a
21:04
specific personalized prediction about your specific illness
21:17
divinatory predictions tend always to be performative so they were a deep awareness that the prediction changes the condition of the future thinks about edibles for example now the prediction
21:29
is there and whatever it does it cannot change the future the prediction changes the future by …
21:51
they are going to propose you to buy a product which you didn't know before
22:09
affect the condition of the future and that's something which in our cultural model culture we know that
22:15
but we are not really well equipped to deal with the uh the performance aspect of relationship would be an annoyance not a …
[…]
23:55
or i definitely agree with you but we are also like um struggling between the different attitudes on the one hand we
24:02
would like to see in advance what happens on the other hand we have to recognize that i mean this or
24:08
now sociological uh central psychological point of view the future does exist yet the future we produced
24:21
circularity in the in the production of the future
[…]
~
TICKNOR, Jonathan L., 2013. A Bayesian regularized artificial neural network for stock market forecasting. Expert Systems with Applications. October 2013. Vol. 40, no. 14, p. 5501–5506. DOI 10.1016/j.eswa.2013.04.013.