Le macchine di Dio. Gli algoritmi predittivi e l'illusione del controllo

by Helga Nowotny, Andrea Daniele Signorelli (Translator), Andrea Prencipe (Contributor)

L'essere umano ha dei nuovi compagni: robot, intelligenze artificiali e non solo. Una nuova specie digitale, con cui stiamo imparando a convivere. Se i nostri timori sono legati al pericolo che gli algoritmi conquistino troppo potere, la vera sfida per il futuro è di evitare di essere noi, inconsapevolmente, a cederglielo, in cambio di qualche comodità in più o di qualche promessa di sicurezza o efficienza. Una grande conquista della modernità è infatti quella di aver "aperto" il futuro, abbandonando una visione predeterminata del nostro destino. Ma cosa succede se iniziamo a fidarci ciecamente delle previsioni degli algoritmi predittivi, a cui affidiamo sempre più responsabilità? Il futuro rischia nuovamente di richiudersi su di noi, dando vita a profezie algoritmiche che siamo noi stessi a far avverare. L'impatto degli algoritmi sulle nostre vite, in questo modo, è destinato a ingigantirsi, con tutto il carico di pregiudizi, errori, imprecisioni e mancanze che si portano dietro. Helga Nowotny ci aiuta a interpretare il futuro che abbiamo di fronte, e a conquistare così la consapevolezza necessaria a non diventare succubi delle macchine di Dio.

197 pages, Paperback

Published June 1, 2022

About the author

Helga Nowotny is President and a founding member of the ERC, the European Research Council. She is Professor emerita of Social Studies of Science, ETH Zurich. In 2007 she was elected ERC Vice President and in March 2010 succeeded Fotis Kafatos as President of the ERC. She holds a Ph.D. in Sociology from Columbia University, NY. and a doctorate in jurisprudence from the University of Vienna.
Her current host institution is the Vienna Science and Technology Fund (WWTF). Helga Nowotny is a member of the University Council of the Ludwig Maximilians University Munich and member of many other international Advisory Boards and selection committees.

Reviews

[Andrei Faragau · Rating 5 out of 5]

A great book that presents in-depth philosophical (and not only) analyses of the impact of predictive algorithms and digitalization on our society. Although it makes for a difficult read at times, the depth of the concepts discussed by the author is impressive.

The analysis that resonated with me the most is that predictions made by data-driven models can lead to the end of the 'open horizon' concept. This concept states that the future has alternative possibilities and the one that materializes is partly influenced by each individual's choices. The 'open horizon' concept has been adopted quite recently (for example, a worker in the past did not believe that through hard work and determination she/he can be part of the elite), and has led many individuals to have optimism about the future. The predictive algorithms can lead to an end to this concept if one blindly believes the prediction it makes and fails to see all alternative possibilities.

Another interesting concept is that the power of predictive algorithms is performative. This means that an algorithm has the capability to make happen what it predicts when the human behaviour follows it prediction. This leads to a false validation of the predictions.

The book also treats subjects such as the increase of identity anxiety with social media, how the narrative of progress that has been with us for the last 200+ years is breaking down, the wisdome that is needed to regulate AI, and work domestication post-COVID.

For the conclusion, part of the author's haiku is perfectly suited, in which she says "future needs wisdom".

[Roger Marc Bond Choquette · Rating 2 out of 5]

First, I've got to get out of the way the fact that the title is no good. But I'm starting a burst of reading on AI and the author's eminence, and the recency of publishing both made this book seem like a good place to start.

I certainly appreciate that Nowotny is intellectually brave--she does not cower from bold positions. Perhaps, though, some of her assertions or analyses need to be defended with more evidence or greater depth. That said, I think more academics should be writing like Nowotny, as a general matter.

The book is perhaps too ambitious--it is not really about Artificial Intelligence, it is about the future generally, and how technology may change its course for societies and individuals. Maybe I was just disappointed by the title, which isn't a very fair criticism.

In any case, Nowotny is obviously very bright and very well read. I like tying books together and the last Eco I read inspired, in part, choosing AI as a topic for several books. So, when she referenced Eco, it did feel satisfying.

Last note: author's commenting on the perils of climate change--caused by others they are criticizing--should refrain from exposing their high-carbon lifestyle. Now I am just being critical, but, I have found personally that this is quite a problem with cosmopolitan academics at the global elite universities--everyone's behavior must change but theirs.

[Ester Perez-Sala · Rating 3 out of 5]

Está bien documentado pero habla de la digitalización en general, no de la IA, partiendo desde el siglo XIX más o menos hasta ahora. Incluye algunas reflexiones de la COVID puesto que se escribió en esa época y es un tema que me da mucha pereza. No es exactamente lo que buscaba, pero ciertos puntos sobre la nueva capa del tiempo digital, el cambio de la noción de privacidad o las diferencias entre robots y humanos llevan a reflexiones interesantes.

[Rebecca · Rating 4 out of 5]

Lettura ampia e interconnessa nei diversi aspetti tecnologici sociali antropologici e normativi che l'introduzione dell'intelligenza artificiale nelle nostre vite implica.
Corposa bibliografia che consente di proseguire gli approfondimenti che più interessano.
Scorrevole come lettura sino al punto che a volte si ha la sensazione che le affermazioni svolte rimangano apodittiche senza una particolare dimostrazione, ma in campo di intelligenza artificiale alcune considerazioni sono necessariamente lasciate alla prova di ciò che vivremo nei prossimi anni.
Consigliato per inserire il tema dell'intelligenza artificiale nel più appropriato quadro complesso in cui opera è smettere di vederla solo come una tecnologia che influisce su alcuni campi limitati o come una distruzione integrale di tutto ciò che siamo.
Utile per riprendere un quadro il più possibile neutro e ragionevole per sapersi meglio muovere nel futuro che stiamo già vivendo o quantomeno comprenderne alcuni meccanismi e linee di sviluppo.

[claudio moderini · Rating 4 out of 5]

AI and Covid

Some very interesting and well sourced assumptions about the risks of trusting predictive tech. The context of social distance and new forms of relations due to Covid are a bit stretched, we are facing a much bigger change of paradigm in how people live, and the question about what is the role of tech programmers in leading the development of AI ecosystems is a bit underrated.