Pagine

venerdì 17 maggio 2019

The digital dimension is a political space

by Enrico Nardelli

(versione italiana qua)

The first commercial computers were sold almost seventy years ago (in 1951, to be precise), and just over thirty years later (in 1982), the designation of the personal computer as "person of the year" marked its widespread adoption and importance. Exactly fifty years ago, in 1969, the "mother of the Internet" (Arpanet) connected the first computers of a handful of American universities into a network, and some twenty years after that (in 1988), the spread of one of the first viruses — even though it had been created to raise awareness of network security issues — brought a still-nascent "network of networks" and the computer systems connected to it to their knees, demonstrating how that infrastructure was at once vulnerable and essential.

Another thirty years have passed, and the explosion (this one genuinely viral) of digital technology is a phenomenon with which we must all, whether we like it or not, come to terms. The digital dimension is now increasingly intertwined with the various social dimensions that govern all the relationships — economic, legal, cultural, and more — established among a group of individuals. As a result, the digital defines a social space that, like every other social space, can be shaped according to different political visions. Around the world, different states have different social systems, which generally reflect in their origins and evolution the collective will and sensibility of their peoples.

It therefore seems to me entirely natural that governments and politicians should also want to exercise their steering and management role with regard to this digital dimension. We should not be surprised or alarmed by this tendency, because people and social relationships have been the same for millennia. We are made of the same stuff and driven by the same desires, aspirations, impulses, and emotions as when we first acquired language, some one hundred thousand years ago.

Following the stringent but appropriate regulation introduced by Europe on personal data through the GDPR, in February of this year India stated very clearly, in the draft national e-commerce policy, that "a nation's data is a collective resource, a national asset" and that "India and its citizens have a right of sovereignty over such data, which cannot be extended to non-Indians".

In Italy, some might cry populism — but I rather doubt that anyone wishing to go to India and exploit its natural resources could do so without being accountable to anyone.

While China — owing to a different political history — was able to bring its digital borders (that is, the exchange points where network traffic enters and exits a country or geographic area) under control in parallel with the spread of the Internet, Russia is now planning to do the same (even though some are portraying this as wanting to "disconnect from the Internet"). Let me be clear: I am a firm and convinced supporter of freedom of speech, and I am absolutely opposed to the censorship that China exercises — and that Russia might exercise — through its control of the Internet.

Even here, however, those who cry scandal should answer the question of whether it is possible to enter the United States — the country where the Internet was born — without presenting valid identity documents. In 2018 I travelled for work to both the USA (which I visit fairly regularly) and Russia, where I was going for the first time. At which border, do you think, were my fingerprints taken? Not at that of the country that now wants to control its digital borders.

These examples confirm that digital space is an integral part of our reality and that it is therefore natural to manage it like any other social structure.

What is needed to this end?

We need to control and govern digital infrastructure and digital data with the wellbeing of the country and its citizens as the paramount concern. Our "digital doubles" are like us — they are exactly us. What would we call rulers who sold their own citizens to foreign powers? We must not allow this to happen in the digital dimension.

We need to understand that digital space, along with the solutions connected to it, must be managed in the interest of the Nation — not of those who, in any case, have virtually unlimited resources to keep acquiring more. We need a politics that returns to exercising, on behalf of the people, the sovereignty that Article 1 of our Constitution assigns to them.

We need a long-term strategic effort for "digital reconstruction." I recall that in the post-war period the State, through IRI (Istituto per la Ricostruzione Industriale = Institute for Industrial Reconstruction), played a fundamental role in the country's development. Its public-private partnership model was admired across Europe as a "third way" for growth, between the market and nationalisation.

In Italy, I fear that the strategic value of digital space has still not been properly understood — people tinker with operational aspects while neglecting to harness it for the country's socio-economic development. Consider, for example, three strategic sectors: schools, the public administration, and industry.

In schools, five years of the National Digital School Plan have produced many promises, a fair amount of spending, but very little real progress — relative to the funding deployed — in terms of genuine advancement in the informatics knowledge of students and teachers. There is still no recognition that training in operational skills alone is not enough: it produces only digital workers, not informed citizens and future leaders.

In the Public Administration, the importance of digital transformation continues to be asserted, without any understanding that generals without adequate armies can achieve nothing. And while I am confident that the "digital soldiers" emerging from our schools and universities — which produce informatics graduates and diploma-holders who distinguish themselves the world over — are well trained, I have some doubt that public sector management is generally equipped to grasp the importance and implications of deploying informatics within organisations.

In the private sector, we still suffer from a cultural deficit that has until now held back the productive system — but for understandable reasons. Our industrial fabric, consisting overwhelmingly of micro, small, and medium-sized enterprises, which made creativity, flexibility, and speed the keys to transforming Italy from a defeated and devastated post-war country into the world's seventh industrial power, rightly sensed — and they were shrewd in doing so — that the systematic use of rigid digital systems would weigh them down intolerably, and they kept their distance. The public spending channelled through the Industry 4.0 initiative is certainly welcome, but if it is used primarily to purchase foreign machinery and solutions, it is hard to see how it can contribute to the country's revival.

Until now, politics has behaved in the way described with masterly precision a few years ago by Morozov — one of the most clear-sighted sociologists in the analysis of digital society — in his book The Net Delusion: "for a mass party today, failing to take responsibility for the digital world is tantamount to failing to take responsibility for the future of democracy itself." The trend is to chase fashionable buzzwords — yesterday the cloud and big data, today artificial intelligence and blockchain. Funding and loans are provided for the acquisition of off-the-shelf digital systems that, six months or a year later, require non-trivial updating costs — instead of investing intensively in the training of in-house personnel capable of contributing to the development of digital solutions and evolving them in step with changing business scenarios. This is an essential element for relaunching the country's economic development, and all the more so as services grow as a share of the economy. (Not to mention the fact that in certain strategic sectors, handing over "the keys to the house" to a stranger is hardly the most intelligent solution!) While in industry the cost of production plant is a proportionally significant item directly linked to production volume, in the delivery of digital services the necessary "means of production" consist essentially of people's brainpower and are largely independent of the number of users. Platforms such as Uber and Airbnb are the most glaring proof of this.

But services built on digital systems can deliver flexibility and adaptability at reasonable cost only if you are able to evolve them in-house, or close to it. This is the only way to build computer systems capable of delivering the productivity gains companies need as they develop. For our productive system, this is the battleground — and it is not a field reserved for the "big players," since informatics, when used well, makes it possible to exploit even very small market niches to good advantage. What is needed, alongside the definition of a development model designed in the interest of the country's growth as a whole — as mentioned above — is a large-scale programme of support for the training of the active workforce, one that goes beyond the necessary but insufficient digital skills of the end user, equipping workers with the conceptual and cultural tools needed for the Italian system to be competitive in this field. The United States understood this very well, going so far as to establish last year the "National Council for the American Worker" with the aim of improving the education and professional training of workers, so as to make them competitive in a global economy that is ever more deeply permeated by digital technology.

Who stands to gain from leaving this state of affairs unchanged? Finding the answers for oneself is the best exercise a citizen who cares about the future of their country can undertake.

--
The original version (in italian) has been published by "Key4Biz" on 14 May 2019.

domenica 5 maggio 2019

Governati da macchine intelligenti? No, grazie!

di Enrico Nardelli

È apparsa recentemente la notizia di un’indagine svolta dal “Center for the Governance of Change” dell’università IE di Madrid (un’università privata in area economico-finanziaria e giuridica) che ha investigato l’atteggiamento dei cittadini europei nei confronti dei cambiamenti tecnologici e come pensano i loro governanti debbano gestire queste trasformazioni.

Il rilevamento, condotto su 2.576 cittadini di 8 nazioni europee, ha riscontrato che tre cittadini su quattro ritengono che, se non sono adeguatamente controllate, le nuove tecnologie causeranno più danni che benefici nel prossimo decennio. E si aspettano quindi che i governi approvino leggi per evitare che la crescente automazione sia causa di perdita di posti di lavoro.

In aggiunta, sette cittadini europei su dieci sono preoccupati che in futuro le persone spendano più tempo socializzando online invece che di persona e ritengono che il sistema educativo non li prepari adeguatamente alle sfide delle nuove tecnologie.

Viene inoltre riportata una certa disillusione verso l’attuale classe politica, tant’è che un cittadino europeo su quattro dichiara che preferirebbe essere governato da un’intelligenza artificiale invece che da politici in carne ed ossa.

Quest’ultimo elemento in sé mi fa un po' rabbrividire: l'idea che qualcuno pensi che le macchine siano meglio degli esseri umani, indica quanto poco le persone in generale siano non dico educate ma almeno informate su cosa sia l’informatica, la disciplina scientifica che è dietro la trasformazione digitale e cosa sia realistico aspettarsi dai sistemi di intelligenza artificiale, uno dei settori dell’informatica in cui in questo momento è concentrata l’attenzione di tutti.

Ed è ancora più rabbrividente il fatto che i laureati siano più favorevoli a questa soluzione rispetto a chi una laurea non ce l’ha. A mio avviso questo può essere spiegato dai recenti risultati della psicologia sociale sulla relazione inversa tra livello culturale e capacità di valutare in modo oggettivo temi politicamente controversi. Tale ricerca ha infatti evidenziato che, mentre persone culturalmente più preparate sono meglio in grado di valutare fatti oggettivi presentati in maniera neutra (per esempio il risultato di un trattamento dermatologico), quando invece le stesse evidenze sperimentali sono presentate in un contesto caratterizzato da un punto di vista politico-sociale (per esempio il risultato di una politica di controllo delle armi private) allora lo stesso gruppo di persone valuta la situazione in maniera peggiore di chi è meno preparato culturalmente.

Da studioso, ho poi delle perplessità su come sono stati riportati i risultati di questa specifica domanda. È stata infatti formulata come "Che ne pensi di consentire ad un’intelligenza artificiale di prendere importanti decisioni su come governare il Paese?” (How do you feel about letting an artificial intelligence make important decisions about the running of the country?) ma si riportano in modo aggregato le risposte “in qualche modo favorevole” (somewhat in favor) e “completamente favorevole” (totally in favor) senza fornirne i valori separati e senza sapere quali siano le altre possibili risposte e con quale frequenza media siano state scelte. Siamo tutti d’accordo, penso, che tra “in qualche modo” e “completamente” c’è una bella differenza.

Ho infine una riflessione importante relativa al pensiero critico e all’educazione ai media, temi che dovrebbero far parte dello studio dell’educazione civica recentemente reintrodotto nelle scuole e la cui proposta sarà discussa in questi giorni alla Camera.

Se ci si riflette con calma e prestando attenzione al punto di vista complementare, si nota infatti che il dato riportato implica che tre cittadini europei su quattro si oppongono a quest’idea di essere governati da “macchine intelligenti”.

Al contrario, il comunicato stampa e il lancio della notizia si sono concentrati su quel cittadino su quattro che vuole essere governato dalle macchine, come – a pensar male si fa peccato ma... – a voler suggerire che invece di una democrazia a suffragio universale che elegge “politici incapaci” siano meglio asettiche “macchine intelligenti” che non sbagliano mai. Mi sembra un’osservazione rilevante sul piano politico e mediatico. “Cui prodest?” e “quis custodiet ipsos custodes?” si studiava un tempo al liceo...

--
Versione originale pubblicata su "Il Fatto Quotidiano" il 30 aprile 2019.