Smart city: surveillance and behavioral economics, the cases of Venice and Ivrea

A reverse translation, based on an excellent work by mrk4m1. Available in English so that even those who do not understand Italian can benefit from it 100% | Matte | Privacy Chronicles

Smart city: surveillance and behavioral economics, the cases of Venice and Ivrea

Author: Matte | Serve una legge | Original Date: 23/04/2022 | Translated by: 31febbraio | Milano Trustless | Link: Original Smart city: sorveglianza ed economia comportamentale, i casi di Venezia e Ivrea

A reverse translation, based on an excellent work by mrk4m1. Available in English so that even those who do not understand Italian can benefit from it 100% and draw their own conclusions, comparing the Italian experience with the reality around them

Today we talk about two different cases but united by the same common thread, that of smart cities and the sudden boom in pervasive systems for surveillance and monitoring people's behavior.

The story begins with a tweet from Venice Mayor Luigi Brugnaro, who was delighted with the restart of the tourist season and the experimentation with online booking that will be mandatory to access the city as a tourist.
Two days later from that tweet in Corriere del Veneto came out another news story, also related to the Venetian tourist season, that caught my attention. The headline was, 'Venice, 20 thousand ghost tourists are in the city but don't result: phone traces are nailing them'.
One passage in the article in particular had interested me:

Abstract from Corriere del Veneto original article (translate below)
Some might think that the Smart Control Room numbers are not accurate, but that is not the case: the system (within privacy rules respect) can see whatever cell phone is in town. And to figure out who is staying in Laguna, they contact the smartphones present at 4 a.m. It can figure out where they are coming from (to say on Easter Monday there were 35,990 residents at home) and where they are. The margin for error exists

The article did not explain what the Smart Control Room they were talking about was, but it did specify that thanks to it the Venice administration is able to see any cell phone in the city, where they come from and where they are.
An issue worth looking into, and so I did.

Venice's 'Smart Control Room'

Apparently, the Smart Control Room (SCR) is a kind of 24-hour operational control room connected to all the city's surveillance and sensor systems. It was developed by Venis (Venezia Informatica e Sistemi SpA) together with TIM and opened in September 2020.
On TIM's website, the SCR is described as follows:

A unified control center, equipped with the latest technologies and equipped to receive information about what is happening not only in the city of Venice, but in the entire metropolitan area. [...]
A huge amount of data and video streams come to the Smart Control Room from the control centers and sensors located throughout the territory. These are, for example, the number of people in Venice, the type of boats in the canals, the passages of road and water public transport, the flow of tourists, the weather forecast and the parking situation. Analyzing them, with the help of processing systems that guarantee privacy [...]

Digging a little more, I then found a video of an event during Milan Digital Week 2021, in which Susanna Jean of TIM spoke precisely about the "Smart Venice" case, describing it as follows (1):

'An innovative project composed of different components and a horizontal layer that collects information in real time, with video analytics, sensors, IoT, and all pre-existing systems. The goal is to give administration and law enforcement a real-time and continuous view of what is happening, including machine learning and data analytics for predictions and simulations. The Cloud layer is concretized in a physical Smart Control Room, which is the place of aggregation of the various components'.

Turning then to the website of Venis SpA, which handled the technology part, I found this information:

In the SCR flows the images of the 400 surveillance cameras, weather forecasts, data on the physical presence in the area, which allows, for example, to identify possible crowds (a critical element especially in this period of emergency) and data on traffic trends on both water and land. [...] All this data is then reprocessed, ensuring privacy, to optimize public services and design new ones, based on scientific data.

To recap: sensors, cameras, Big Data, IoT and machine learning. All of it, packaged and available for use by administrators and law enforcement in the Smart Control Room.
Both TIM and Venis are keen to emphasize that it is all done 'while ensuring privacy is respected'. Yet, aside from marketing claims, I could find no concrete assurances.
Nowhere could I find any information, even basic information, about the processing of personal data; just as I could not find any information regarding guarantees for the rights and freedoms of people who find themselves under 24-hour surveillance by such a control room.
Someone might say, "but you know Matte, it's aggregated data, so it's anonymous, that's how privacy is protected."

Not really.

Even assuming, as is plausible, that only aggregate data reaches the SCR operators, that data is the result of processing that is done upstream (by TIM and/or other parties) from personal data and metadata (2) obtained thanks to sensors, cameras, and telephone networks in the city.
If we then want to rely on what the Corriere del Veneto wrote, we certainly cannot deny that we are in the presence of a very pervasive processing of personal data: 'the system manages to see any cell phone in the city [...] smartphones are counted at 4 a.m.. You can figure out where they come from and where they are'.
There is no doubt that the one carried out in Venice is a treatment that involves the systematic use of huge amounts of data for the observation, monitoring and control of people, as well as a treatment of metadata for organizational and 'security' purposes related to the city.
Since the processing is done on behalf of the Municipality, this must ensure respect for the rights and freedoms of citizens and mitigate the risks arising from the processing of data - as also required by European law.
Where are transparency, proportionality, legitimacy, and risk assessment? Who knows?
This particular type of data processing also falls under the high-risk categories identified by the Privacy Authority in 2018, for which an impact analysis is mandatory:

Privacy Authority rules for mandatory impact evaluation (translate below)
Processing involving the systematic use of people's data for the observation, monitoring or control, including the collection of data through networks, also carried out on.line or through apps, as well as the processing of unique identifiers capable of identifying users of information of a service society including web services, interactive TV, etc., with respect to usage habits and viewing data over extended periods. This also includes processing of metadata e.g. in telecommunications, banking, etc., carried out not only for profiling, but more generally for organizational reasons, budget estimates, technology upgrades, network improvements, anti-fraud service offerings, anti-spam, security, etc.

What about the security issue, which is taken up many times by all press releases? How has the positive impact of this pervasive surveillance on citizen security been assessed? More importantly - security from what?

As I also pointed out in this article, Italian politics for years has been pivoting on the perception of security (or insecurity) to justify the increasingly intensive use of physical surveillance technologies. But beyond perception, there is nothing. In fact: the data say the opposite: crime rates usually have nothing to do with the amount of cameras in the city.

Reservations and control rooms

To close the circle on Venice, I finally link back to the news story about the experimentation with reservations to access the city.
What happens when we combine reservations referable to identified people with a control system like SCR's? Well, what happens is that we have all the preconditions for Venice to become an open-air cage.
The reservation system will allow the city government to identify every single person visiting Venice, while the SCR will allow it to monitor their movements and accommodation. Cross-checking data as needed will be a breeze.

Are tourists (Italian and international) aware that they will be under continuous surveillance from the moment they set first foot in Venice until the moment they leave? What about the residents?

The "Smart Ivrea" experiment

While researching the Smart Control Room, I came across another project in which TIM participated, that of 'Smart Ivrea'.
The 'Smart Ivrea' project is an initiative led by AgID and funded by MISE. It is also the first prototype for testing a national platform for smart community management (3).

The trial began in the first quarter of 2020 and is now almost at its conclusion. The idea is to replicate and scale the platform nationwide, based on the results of the Ivrea trial.
The objectives of the platform are described in the MISE funding documentation:

The novelty consists in the development of a Smart cities-as a service (Scaas) model, aimed at optimizing the delivery of existing public services by introducing some principles of behavioral economics (reward system for citizen's assumption of virtuous behavior, sentiment analysis) and participatory governance (eVoting and crowdfunding), determining active citizen participation in the social, cultural and political life of the territory, going as far as launching the first national virtual currency ecosystem (Ivrea-Coin), through which citizens can purchase both services provided by the administration and, eventually, those offered by SMEs.

Investigating how the platform works, I found an interview with Marco Pittorri of Trust Technologies (TIM partner), in which he describes the system (4) as follows:

Citizens can access services via app. The user authenticates with a SPID-certified identity, which is then taken to blockchain and associated with an ID wallet used by the citizen with two main functions: to pay for public services (buses, taxes, etc.) and to receive back a reward according to the correct performance of his role as a citizen: by paying taxes regularly, using public services or buying on affiliated stores to enhance local economy, he is able to get back IVREA COIN, which he can reuse to pay for services of the municipality. The benefit to citizens is participation in the life of the municipality and the rewarding of good behavior.

The core of the project revolves around the concept of behavioral economics and Ivrea Coins, which incentivize citizens to perform behaviors deemed correct by those who developed and implemented the system.

Workflow to obtain Ivrea Coins

Nudging and hypernudging

The topic of behavioral economics is closely related to that of 'nudge theoryì developed by Thaler and Sunstein in 2008. They defined 'nudge' thus:

A nudge, as we will use the term, is any aspect of the choice architecture that alters people's behavior in a predictable way without forbidding any options or significantly changing their economic incentives.

From their definition we understand two things: nudge is the result of a choice made by those who think about and develop a particular system, which can be the supermarket shelf as well as an app related to the smart city. It is a mechanism that, without prohibiting or forcing people toward certain choices, manages to alter their behavior in a predictable way.
In essence, nudge is a mechanism for standardizing human behavior and making it predictable.
When incentives are linked to more pervasive digital systems, with artificial intelligence and Big Data technologies that can simulate and predict human behavior, the issue becomes quite complicated.

In these cases, some speak as hypernudging.

Machine learning algorithms are often used to process data about people's decisions and actions, as is the case, for example, on social networks, creating the so-called 'filter bubble' effect: the algorithm collects data about our interactions and proposes content to us based on our past actions. Thus a retroactive vicious cycle is established, in which people are prompted to take certain actions based on the choices they made previously.
The hypernudging effect leads to insidious coercion: what appears to be a choice is actually a defined set of standardized options created by the system. The opposite of free will and freedom of self-determination.
The Ivrea experiment probably does not yet reach this kind of capability, but the road is certainly marked.

The denial of what makes us human

In 1998 James C. Scott (5) argued that any attempt by the state to rationalize, simplify and standardize society has always led to the creation of authoritarian powers, which often vent into genuine totalitarianism and human tragedies.

China was perhaps among the first countries in the world to theorize on these issues, with Lin Junyue's famous paper mentioned here. The purpose was to incentivize individuals to perform certain behaviors to shape and rationalize society as a whole.
As I often say, the problem with incentives is precisely that they work.

In the article 'Point Citizenship and the Ethical State, from Rome to Bologna', I was saying that social scoring systems are the enactment of the Ethical State, in which the citizen becomes a cog in the system, with no real freedom of choice.

It is no coincidence that we always talk about 'virtuous' or 'correct' behavior: these are necessary words to reinforce the idea of the righteousness and dignity of people only within the community, made up of people who behave 'correctly', according to standards predetermined by third parties. Those who step outside the standards remain outside the community, thus losing all dignity as citizens. In part, we have seen this with the Green Pass.

Social scoring/premium systems are the rejection of all individualism, critical thinking and behavior diverging from the standard. Human evolution is based on (moral) action and experimentation with the unknown. Algorithms, on the other hand, work in reverse: they reason only on the basis of the past background. A human society governed by algorithms is an immobile human society, devoid of morality and incapable of evolution. Basically, the denial of everything that makes us human beings.

I do not think that Italian mayors make such reasoning. Rather, they get charmed by projects pushed by technocrats in search of glory and public funds, in a historical period where artificial intelligence, IoT and blockchain are the buzzwords to get all kinds of funding.
They sell these systems with fine words, presentations and catchy press releases. They tell us that they are for our security and public efficiency. That innovation is beautiful. That people will be at the center of ecosystem of services.

My feeling is that, yes, humans will increasingly be at the center -- but of a hi-tech corral; like cattle grazing, waiting for slaughter.

1 Paraphrased text, link to interview
2 Metadata refers to that information that is not directly referable to individuals, but describes events, actions, or other information. A typical example of metadata is location data (longitutine and latitude of a device at a given time).
4 Paraphrased text, link to interview
5 Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed

Se non diversamente specificato, il contenuto di questo articolo è concesso in licenza ai sensi del CC BY-SA 4.0