February 2020 - nexa - server-nexa.polito.it

Re: [nexa] "A European strategy for data "

by M. Fioretti

Due domande per G. Attardi: On Wed, Feb 26, 2020 07:23:01 AM +0100, Andrea Glorioso wrote: > On Tue, Feb 25, 2020 at 3:12 PM Giuseppe Attardi <[1]attardi(a)di.unipi.it> > ... > È per questo che secondo me bisognerebbe ragionare sui dati collettivi > come di un Common Good, la cui proprietà non è dei singoli, ma della > comunità. a monte di questo, bisognerebbe ragionare sul fatto che certi insiemi di dati collettivi non dovrebbero proprio esistere/essere creati, perche' dopo sarebbe difficilissimo impedirne deanonimizzazione (correlandoli con altri insiemi) o altri abusi, no? > Chiunque accoglie dati da cittadini o da fonti pubbliche, > dovrebbe non esserne proprietario ed essere obbligato a > renderli accessibili a chiunque a condizioni eque. Qui dopo "accoglie dati" manca almeno un "non sensibili", no? Marco -- M. Fioretti http://mfioretti.com http://stop.zona-m.net Your own civil rights and the quality of your life heavily depend on how software is used *around* you

Feb. 26, 2020

Re: [nexa] "A European strategy for data "

by Andrea Glorioso

Salve, On Tue, Feb 25, 2020 at 3:12 PM Giuseppe Attardi <attardi(a)di.unipi.it> wrote: > Il 19/02/20 23:38, J.C. DE MARTIN ha scritto: > > COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE > COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF > THE REGIONS > > A European strategy for data > > > Documento (PDF, 35 pagin) in allegato. > > juan carlos > > > Today 80% of the processing and analysis of data takes place in data > centres and centralised computing facilities, and 20% in smart connected > objects, such as cars, home appliances or manufacturing robots, and in computing > facilities close to the user (‘edge computing’). By 2025 these > proportions are likely to be inverted. > > Ammesso e non concesso che questo sia vero, a questo punto l’obiettivo di > > The aim is to create a single European data space – a genuine single > market for data, open to data from across the world > sarebbe inutile o poco rilevante. > Ciascun dispositivo userebbe i propri dati e dello spazio singolo europeo > ne potrebbe fare a meno. > Uno degli scopi dello spazio singolo europeo dei dati è assicurare l’interoperabilità legale e tecnica tra i dispositivi. Che poi tali dispositivi (o meglio, i relativi progettisti e utenti) vogliano sfruttare tale possibilità o meno, è una scelta. In realtà cedo stia sfuggendo il fatto che sono proprio le raccolte di dati > centralizzate quelle che costituiscono un valore economico e un vantaggio > competitivo. > I dati di singoli hanno ben poco valore, solo raccolti tutti insieme in > dataset collettivi hanno un valore, quanto meno per ragioni di statistica. > Ti posso garantire che, almeno all’interno della Commissione Europea, tale fatto non sfugge. È per questo che secondo me bisognerebbe ragionare sui dati collettivi come > di un Common Good, la cui proprietà non è dei singoli, ma della comunità. > Chiunque accoglie dati da cittadini o da fonti pubbliche, dovrebbe non > esserne proprietario ed essere obbligato a renderli accessibili a chiunque > a condizioni eque. > È la quinta delle mie re4gole dell’economia robotizzata. > Quali sono le altre regole? Ciao, Andrea -- -- Andrea Glorioso Twitter: @andreaglorioso Facebook: https://www.facebook.com/andrea.glorioso LinkedIn: http://www.linkedin.com/profile/view?id=1749288&trk=tab_pro

Feb. 26, 2020

Cambiare il mondo con un click: una riflessione sul clicktivism

by don Luca Peyron

Semplice ed interessante panoramica sull'attivismo on line Buona lettura dl http://www.beunsocial.it/cambiare-il-mondo-con-un-click-una-riflessione-sul… _________________________ www.apostolatodigitale.it via XX settembre 83, Torino tel. 011 5156239

Feb. 25, 2020

Re: [nexa] "A European strategy for data "

by Giuseppe Attardi

Il 19/02/20 23:38, J.C. DE MARTIN ha scritto: > COMMUNICATION FROM THE COMMISSION TO THE EUROPEAN PARLIAMENT, THE > COUNCIL, THE EUROPEAN ECONOMIC AND SOCIAL COMMITTEE AND THE COMMITTEE OF > THE REGIONS > > A European strategy for data > > > Documento (PDF, 35 pagin) in allegato. > > juan carlos Today 80% of the processing and analysis of data takes place in data centres and centralised computing facilities, and 20% in smart connected objects, such as cars, home appliances or manufacturing robots, and in computing facilities close to the user (‘edge computing’). By 2025 these proportions are likely to be inverted. Ammesso e non concesso che questo sia vero, a questo punto l’obiettivo di The aim is to create a single European data space – a genuine single market for data, open to data from across the world sarebbe inutile o poco rilevante. Ciascun dispositivo userebbe i propri dati e dello spazio singolo europeo ne potrebbe fare a meno. In realtà cedo stia sfuggendo il fatto che sono proprio le raccolte di dati centralizzate quelle che costituiscono un valore economico e un vantaggio competitivo. I dati di singoli hanno ben poco valore, solo raccolti tutti insieme in dataset collettivi hanno un valore, quanto meno per ragioni di statistica. È per questo che secondo me bisognerebbe ragionare sui dati collettivi come di un Common Good, la cui proprietà non è dei singoli, ma della comunità. Chiunque accoglie dati da cittadini o da fonti pubbliche, dovrebbe non esserne proprietario ed essere obbligato a renderli accessibili a chiunque a condizioni eque. È la quinta delle mie re4gole dell’economia robotizzata. — Beppe

Feb. 25, 2020

Re: [nexa] Artificial intelligence yields new antibiotic

by Roberto Resoli

Il 24/02/20 21:12, don Luca Peyron ha scritto: > > > http://news.mit.edu/2020/artificial-intelligence-identifies-new-antibiotic-… Era anche su TG3 Leonardo di Venerdì scorso 21/02 https://www.rainews.it/tgr/rubriche/leonardo/index.html?/tgr/video/2020/02/… rob

Feb. 25, 2020

Re: [nexa] L’intelligenza artificiale minaccia la nostra responsabilità morale: ecco come

by cristina

Grazie per la segnalazione.Argomenti importanti sui quali si dovrebbe riflettere ed avere più consapevolezza.Cari saluti.C. Inviato da smartphone Samsung Galaxy. -------- Messaggio originale --------Da: Alberto Cammozzo <ac+nexa(a)zeromx.net> Data: 24/02/20 10:31 (GMT+01:00) A: nexa(a)server-nexa.polito.it Oggetto: Re: [nexa] L’intelligenza artificiale minaccia la nostra responsabilità morale: ecco come Grazie per aver condiviso l'interessante articolo: sono il genere di questioni sull'impatto profondo delle tecnologie che dovremmo porci di più. Nel 2008 a e-Privacy presentai una breve rifessione sulla sostituzione delle norme giuridiche con vincoli tecnici, ispirata dalla lettura di un saggio di Joseph Raz. La deresponsabilizzazione è solo uno degli aspetti emergenti dal sostituire il "quod licet" delle _ragioni_ che stanno dietro alle norme con il "quod potest" dei vincoli tecnici. A chi interessasse il documento è qui: <http://cammozzo.com/Papers/Cammozzo-RagioniVincoliNorme.pdf> Un saluto dalla zona di interdizione. Alberto On 22/02/2020 23:29, don Luca Peyron wrote: > Un articolo interessante che mi pare non sia ancora stato qui condiviso. > > Buona lettura > > dl > > https://www.agendadigitale.eu/cultura-digitale/lintelligenza-artificiale-mi… > > _________________________ > www.universitari.to.it <http://www.universitari.to.it> > via XX settembre 83, Torino > tel. 011 5156239 > > _______________________________________________ > nexa mailing list > nexa(a)server-nexa.polito.it > https://server-nexa.polito.it/cgi-bin/mailman/listinfo/nexa _______________________________________________ nexa mailing list nexa(a)server-nexa.polito.it https://server-nexa.polito.it/cgi-bin/mailman/listinfo/nexa

Feb. 24, 2020

Artificial intelligence yields new antibiotic

by don Luca Peyron

http://news.mit.edu/2020/artificial-intelligence-identifies-new-antibiotic-…

Feb. 24, 2020

Éric Sadin: stiamo in guardia dalla tecnologia antiumana e dalla data-driven society

by don Luca Peyron

Invito alla lettura dl https://www.morningfuture.com/it/article/2020/02/24/eric-sadin-tecnologia-s… _________________________ www.universitari.to.it via XX settembre 83, Torino tel. 011 5156239

Feb. 24, 2020

EU Commission Embraces Remote Facial Recognition? :: Civil Liberties Union for Europe

by Alberto Cammozzo

<https://www.liberties.eu/en/news/ec-white-paper-on-ai-and-facial-recognitio…> The latest version of the European Commission's white paper on AI does not specify a temporary ban on remote facial recognition. Today, the European Commission presented its white paper on AI. In line with what the media found out last week, the released version - as opposed to an earlier draft - does not suggest a temporary ban on remote facial recognition. During a press event last Thursday, Commission Vice-President for Digital Margrethe Vestager said that the Commission would not propose any EU-level regulatory measure now. “So what we will say in the paper in a very lawyered up language is, let’s pause and figure out if there are any [situations], and if any, under what circumstances facial recognition remotely should be authorised," Vestager said. At the end of the day, this means that law enforcement agencies all over Europe will increasingly use the technology in ways they see fit, in many cases under highly questionable interpretations of the national laws. (The best research on this was conducted by Privacy International in the UK last year. If you are interested in how UK police are breaching the rights of our former fellow-citizens, read about their project here.) Police surveillance coupled with facial recognition technology endangers our democracies. It hinders our right to speak our minds, to meet others and publicly express our disagreements with people in power. It is especially dangerous in times when authoritarianism is on the rise. The white paper on AI will be the starting point of a 12-week consultation. Liberties strongly hopes that in those 12 weeks the Commission will listen to the human rights community’s concerns, and its legislative proposal, expected to be put forward late this year, will appropriately protect our civil liberties and human rights.

Feb. 24, 2020

EU Police Push for Pan-European Facial Recognition Network

by Alberto Cammozzo

<https://theintercept.com/2020/02/21/eu-facial-recognition-database/> A police investigator in Spain is trying to solve a crime, but she only has an image of a suspect’s face, caught by a nearby security camera. European police have long had access to fingerprint and DNA databases throughout the 27 countries of the European Union and, in certain cases, the United States. But soon, that investigator may be able to also search a network of police face databases spanning the whole of Europe and the U.S. According to leaked internal European Union documents, the EU could soon be creating a network of national police facial recognition databases. A report drawn up by the national police forces of 10 EU member states, led by Austria, calls for the introduction of EU legislation to introduce and interconnect such databases in every member state. The report, which The Intercept obtained from a European official who is concerned about the network’s development, was circulated among EU and national officials in November 2019. If previous data-sharing arrangements are a guide, the new facial recognition network will likely be connected to similar databases in the U.S., creating what privacy researchers are calling a massive transatlantic consolidation of biometric data. The report was produced as part of discussions on expanding the Prüm system, an EU-wide initiative connecting DNA, fingerprint, and vehicle registration databases for mutual searching. A similar system exists between the U.S. and any country that is part of the Visa Waiver Program, which includes the majority of EU countries; bilateral agreements allow U.S. and European agencies to access one another’s fingerprint and DNA databases. “This is concerning on a national level and on a European level, especially as some EU countries veer towards more authoritarian governments.” Although new legislation following the report’s recommendation is not yet on the table, preparatory work is ongoing. Information provided by the European Commission to the European Parliament last November shows that almost 700,000 euros (about $750,000) are going to a study by consultancy firm Deloitte on possible changes to the Prüm system, with one part of the work looking at facial recognition technology. The European Commission has also, separately, paid 500,000 euros to a consortium of public agencies led by the Estonian Forensic Science Institute to “map the current situation of facial recognition in criminal investigations in all EU Member States,” with the aim of moving “towards the possible exchange of facial data,” according to a project presentation sent to national representatives in Brussels. “This is concerning on a national level and on a European level, especially as some EU countries veer towards more authoritarian governments,” said Edin Omanovic, advocacy director for Privacy International. Omanovic worries about a pan-European face database being used for “politically motivated surveillance” and not just standard police work. The possibility of pervasive, unjustified, or illegal surveillance is one of many critiques of facial recognition technology. Another is that it is notoriously inaccurate, particularly for people of color. “Without the transparency and legal safeguards for facial recognition technology to be lawful,” said Omanovic, “there should be a moratorium on it.” The EU has taken big steps to connect a host of migration and security databases in recent years. New legislation passed last April established a database that will hold the fingerprints, facial images, and other personal data of up to 300 million non-EU nationals, merging data from five separate systems. According to the report by 10 police forces, Deloitte consultants proposed doing the same with police facial images, but the idea was met with unanimous opposition from law enforcement officials. Nonetheless, the report recommends linking all of EU member states’ facial databases, which would seem to have the same practical effect. In another internal EU police report — this one from a working group on Prüm that looked at the exchange of drivers’ license data — police note that “a network of interconnected national registers can be regarded as a virtual European register.” To the police, the advantages of interlinked facial recognition databases are clear. The Austria-led report views the technology as a “highly suitable” biometric tool for identifying unknown suspects and suggests that the databases should be created and linked “as quickly as possible.” It also recognizes the need for data protection safeguards, such as human verification of any automated matches, but privacy experts argue that the creation of any such system is the first step toward greater sharing and linking of data where such controls are inadequate. European moves to consolidate police facial recognition data closely resembles similar efforts in the U.S., said Neema Singh Guliani, senior legislative counsel at the American Civil Liberties Union. Many U.S. law enforcement agencies work out of “fusion centers,” where they are co-located and able to share data. If you have an information-sharing agreement with the FBI or the Department of Homeland Security, said Guliani, “there’s a risk that functionally the information may be shared with additional levels of U.S. law enforcement.” “It raises many questions,” she added. “How police are using facial recognition and gathering images, as well as in the U.S. with regard to due process and First Amendment expression. Given existing information sharing relationships, it’s very likely that the U.S. would want access to that information.” As far back as 2004, the U.S. Embassy in Brussels was calling for a relationship with the EU that allowed “expansive exchanges and sharing all forms of data, including personal data.” In recent years, efforts toward that goal have intensified. According to a Government Accountability Office report, in 2015, the Department of Homeland Security began demanding the implementation of the data-sharing agreements required of Visa Waiver Program countries. This has included the FBI providing assistance to other states to set up the necessary computer networks. Austria, to take one example, began checking fingerprints against the FBI’s criminal fingerprint databases in October 2017, explained Reinhard Schmid, a senior official in the Austrian criminal intelligence service. Since then, about 12,000 individuals’ prints have been cross-checked, leading to 150 matches. “Around 20 of these identified persons were under investigation and suspected of membership of terrorist organizations,” while in 56 cases individuals had attempted to use a false identity, said Schmid. “Their logic here is, ‘When I have a serious crime and I want to run someone’s photo against a database, why shouldn’t I have this?’” said Guliani. Yet, she added, the privacy implications were enormous. “Once you have the access, you ultimately have the ability to identify almost anyone, anywhere.” The report by 10 police forces calls for Europol, the EU agency for police information and intelligence sharing, to play a role in exchanging facial recognition and other biometric data with non-EU states. This echoes recommendations from European governments themselves: A July 2018 declaration called for the commission to consider “broadening the scope” of the Prüm network and for Europol to take the lead on data sharing with third countries. The FBI and Europol did not respond to questions about data-sharing agreements between the EU and the U.S. A spokesperson for the European Commission acknowledged the prospect of adding facial recognition data to the Prüm network, but declined to go into more detail.

Feb. 24, 2020