Is contact tracing a privacy hazard?

Contact tracing is a very popular word these days, mainly because apps designed to keep track of Covid-19 trends rely on this technology. Globally, there are many countries already using apps of this type, and many more are being developed, including Italy’s Immuni. In our country from the moment the app was announced there has been a fierce debate between those in favor and those against the use of this solution. Impossible at the moment to say who is wrong and who is right, and as often happens in these cases the truth probably lies in the middle. Those against see the use of these apps as a heavy invasion of privacy and point to it as the first step that will limit our freedoms. On the other hand, those in favor highlight the usefulness of these apps in keeping contagious people in check, then pointing out that if most of us give up personal data for free to Google, Facebook and other tech giants (who exploit it for advertising profiling), why bother with those provided for a common good.

So many solutions

At the global and European level there is a lot of confusion in this field and each country has made its own choices. There are two different approaches in the development of these applications, those of pure contact tracing (referred to as centralized apps) that exploit the guidelines of the PEPP-PT (Pan-European Privacy-Preserving Proximity Tracing) consortium and those of exposure notification (referred to as decentralized apps) that exploit the DP-3T model provided by Apple and Google. In simple terms for the uninitiated, the main difference is that centralized apps communicate data to a remote server, while in decentralized apps the data remains on the device that communicates via Bluetooth standards. In Italy, the government has chosen the Immuni app developed by the experts at Bending Spoon, which in principle can be called a hybrid model between the two technologies. It should be pointed out that since it is still under development, the app has been modified toward the use of the more privacy-secure decentralized model. But of course a full evaluation on this app can only be made when the final version is made available.

How does it work?

The Immuni app will apparently be able to be used on a completely voluntary basis and is designed to track Covid-19 infections. According to current information, there will be a citizen medical record section within the app that will be activated at a later date. To do the tracking, the app will take advantage of Bluetooth technology so that it will anonymously record the people we have come in contact with, sending an alert if these people test positive for infection. Basically, the app will store on the smartphone the codes of the bluetooth devices one has come in contact with, while anonymity will be guaranteed by encryption systems that prevent the code from being associated with the identity of the owner of that device. If a citizen, after a coronavirus test, tests positive, he or she will be able to upload the collected data to a server via the app. This will take care of sending a notification to the devices of people who by proximity and time of contact with the sick person are potentially at risk of infection. The notification will contain a message decided by health authorities that will ask people to follow a particular protocol, such as contacting a number to perform a swab test, or remain in isolation.

Benefits and risks

As we can guess in order to preserve the privacy of citizens, the system devised is undoubtedly complex and, if all goes well, will undoubtedly be useful. But will Immuni really be able to work? It is estimated that to be effective, the app will have to be used by at least 60-70 percent of the population, which is not an easy goal to achieve on a voluntary basis; considering the people who will not want to install it and those who will not have an adequate phone available for use. A social assessment must then be made, whether the app will be affectively anonymous in how many, in the case of receiving an alert notification, will implement the advice in it. To all this must be added the risks of data breach, that is, a security breach in which sensitive, protected or confidential data are accessed, a danger not so much for Bluetooth detections (for this the issues are different), but for communications from the central server that will send the alert messages. All this is without considering that many of the smartphones on which the apps will run will not have up-to-date operating systems with the latest security updates and some will already be compromised by malware. If even the most emblazoned hi-tech giants succumb under the blows of hackers, how is it possible to think that a personal app, active on smartphones of all kinds and partly connected to the public administration can be completely secure? Perhaps the greatest danger lies in the citizen’s medical records that will be stored on smartphones, devices that are certainly easier to hack. After all, the Bluetooth standard in both Android and Apple phones over the years has manifested fairly frequent security flaws.