Home / Blog / Post-Pandemic Scenarios – III

Post-Pandemic Scenarios – III

Facial recognition at the Hong Kong Airport. The image picked up by the videocamera at the gate is matched with the one stored in the passport chip to check the person’s identity. Image credit: Hong Kong Airport Authority

The pervasiveness of Digital Signature

The second scenario presented in the FTI’s report is focussing on Scoring and Recognition. It is interesting to notice that the very first line of this scenario states:

Anonymity is dead

Indeed, in spite of all the talks about privacy, the reality is that each one of us has, most of the time unwittingly, a digital identity in the cyberspace that can be matched in various ways to our physical persona. There may, actually, be several digital identities for each of us. The chip on my passport has biometric information that can be used to check my identity at an automated boarding gate at the airport or at an international border, my fingerprints are recognised by my smartphone and possibly by my car lock, my face is recognised to enter a protected area at the office, Alexa recognises my voice… Are these separate instances of identity, are the data used in each instance confined in the specific application domain? I, for one, can trust that the fingertips identity recognition on my car door keeps the data to itself, locally, and that it may be unlikely to be hacked, I have the same trust about my smartphone keeping my biometric data locally but I am a bit less confident that there won’t be any hacking attempt on those data and I am definitely less sure about the way Alexa (and the likes) are managing my identity data. In the end I am trading my confidence versus my convenience of using those devices/apps/services.

As noted, there are several ways of creating, and then checking, the identity of a person and once you have that someone can start storing (tracking) information on your whereabouts, learn about your behaviour, make judgement and pick up judgment from others, in other words give you a “score”. This is already happening (in on line commerce, in social networks, in social relations in China…) and it is going to become the norm by the end of this decade. Hence the reason why FTI has aggregated the identity recognition part with the scoring part…. and they both may be convenient and frightening.

Recognition technology is bound to evolve rapidly, actually according to the FTI report the evolution will be faster than the regulatory effort to provide safeguards (hence, I guess, the tagline “Anonymity is dead”).

Facial recognition, as an example, has already reached a very high sophistication and, most importantly, it has become affordable and practical. China has experimented (is using?) a 500 Mpixel camera that is able to pick up thousands of faces (at a stadium, at a city crossing…) with sufficient clarity, even in low light condition, to allow a software to perform facial recognition for thousands of people at a time. If only a few years ago facial recognition had a bias for certain races (because of the images used to train the system) now it can recognise any face, even a partial face. It can even recognise cartoon characters in movies and pets’ faces. we can only expect it to become even more effective in the coming years.

It is not just facial recognition. Voice recognition has made significant progress (and the Covid-19 further stimulated studies on identifying voice patterns that could be associated with infection) to the point of recognising accent thus pinpointing the origin of a person, even if that person is speaking a foreign language, as well as to detect emotion. This very same applications that are able to detect subtle nuances in voices are also able to duplicate them, thus fuelling deep fake issues.

By coupling different sources of identity recognition, like voice, messages sent on social networks, web sites and the way that perrson is looking at information on a given site, AI software can reconstruct the “personality” of that person and predict what would be the reaction to certain stimuli (information, images…).

This multiple sourcing of data obtained through “recognition” technologies applied to a person is summed up in the report by the tagline:

Your body is a dataland

This is seen as an enabler, and a driver towards tele-healthcare, an area that has taken up steam as consequence of the pandemic.

Recognition technology is further progressing through computational photography where there is a lot of work, and results, in recognising objects in an image.

All of the above can be summarised under the trend towards a diffuse, pervasive digital signature of people and objects that will create a mirroring of any entity in the cyberspace.

Signal processing, the technology underpinning all kinds of recognition, has progressed enormously. As an example it is now possible to detect what is going on behind a wall, in an apartment as an example, by analysing the subtle changes in the WiFi field generated by an access point inside the apartment and trickling out of it (a person moving inside the apartment creates small perturbations to the electromagnetic field and these perturbations can be detected and analysed – watch the clip, courtesy of  MIT researchers). It is also possible to identify a person by analysing the distortion caused to sound waves as they pass through that person body (bio-acustic signature).  There are many more ways to pick up biometric parameters that can uniquely identify a person and, as you can see, many of these technology are based on widely available signals so that identification can occur everywhere, undetected by the identified people.

More to follow….


About Roberto Saracco

Roberto Saracco fell in love with technology and its implications long time ago. His background is in math and computer science. Until April 2017 he led the EIT Digital Italian Node and then was head of the Industrial Doctoral School of EIT Digital up to September 2018. Previously, up to December 2011 he was the Director of the Telecom Italia Future Centre in Venice, looking at the interplay of technology evolution, economics and society. At the turn of the century he led a World Bank-Infodev project to stimulate entrepreneurship in Latin America. He is a senior member of IEEE where he leads the Industry Advisory Board within the Future Directions Committee and co-chairs the Digital Reality Initiative. He teaches a Master course on Technology Forecasting and Market impact at the University of Trento. He has published over 100 papers in journals and magazines and 14 books.


  1. giancarlo micono

    Sono una persona matura; per passione ed esperienza collaboro con un editore di testate prevalentemente on-line nella provincia torinese. Questa Città e Regione hanno una storia di “riciclaggi”, da ex-capitale politica a ex patria nazionale di radio e cinema, oltre all’auto. Di AI mi interessano molto gli aspetti socio-culturali, il cambiamento di atteggiamento richiesto fra la gente (Society 5.0 in Giappone). Il Centro AI a Torino potrebbe essere l’inizio dell’ennesima evoluzione. Mi sembra il ns pezzo forte in questo momento, piacerebbe scriverne, mi piacerebbe che se ne parlasse di più….