What does it mean to be a person? What shapes our identity? Are we a collection of continuous memories and experiences? Do we have just one identity, or do multiple personas coexist? If you’ve come across “Severance”, these and other questions have probably crossed your mind. Created by Dan Erickson, the series tells the story of employees of a company who have two totally separate identities – personal and professional -, and where memories are not shared. From fiction to reality, there are many questions about the control we have over the different “versions” of our identity. In a world where digital documents and verifiable credentials are increasingly relevant, can we lose privacy and autonomy when it comes to sharing our data?
Now: imagine yourself, walking down the street. You’re approached by a police officer who asks you to identify yourself. Your first instinct is to show your ID card. But isn’t a person’s identity more than a mere document with personal information?
According to the World Bank’s Identification for Development Initiative, close to 850 million people worldwide do not hold an official ID; also, 3.3 billion do not have access to digital identification. This lack of documentation primarily affects people in sub-Saharan Africa and South Asia, with more than half being children whose births have not been recorded.
The IDINA project emerged in this context, aimed at mapping trust-based social relationships as the foundation for an identification system involving schools, healthcare institutions, NGOs, and local authorities as sources of information about citizens’ births and life events. A bold concept that earned them the first prize during the third edition of the IN3+ award.
Over several years, INESC TEC researchers have had the opportunity to work in different regions – particularly in some Portuguese-speaking African countries (PALOP) and Timor-Leste -, where they found evidence of a problem affecting many people: the difficulty they face in obtaining official identification documents. According to Vítor Fonte, this was an opportunity to realise how far we are from the reality of these countries: “we began to outline a pragmatic solution for the identification of people in regions where the State struggles to ensure civil registration. This is one of the best examples of technology applied to the real world and people’s needs. Fundamental research is crucial, but projects like this are key to advance science, showing the relevance of our work inside our institution.”
João Marco, another researcher involved in the genesis of this project, explained that the entire system was built based on international technical standards currently used in official civil registries. This means that, in addition to safeguarding security and privacy issues, it also provides for the interoperability of the system. Since the target regions are often the field of work of non-governmental organisations, it would be useful to try to map the relationships of trust that these entities have with the populations and allow the creation of a pragmatic system of citizens’ identification. “We needed a different approach from what is typically adopted, developing a system that, although supported by digital infrastructures, did not require the same from the citizens. The idea was that, in many of the regions where people still do not have official identification documents, they could be practically recognised by the multiple entities they interact with daily. These entities – which, in many cases, complement or even replace the State in the services provided to the populations – will, in turn, be able to know the population better, coordinate and plan more effectively and efficiently the services provided,” he said.
What is the truth?
How to ensure the reliability of the information provided by these entities? “IDINA focuses on the creation of a non-authoritative identity, i.e., one that is not officially recognised by the State. Instead, it relies on a network of entities that, over time, contribute to the creation of a reliable profile about individuals. Researcher João Marco stated that “if we think about it, our identity documents do not necessarily reflect the truth; I know some people who have an incorrect date of birth. With IDINA, as multiple entities provide data about the same individual, the information is improved, confirmed or corrected, and we attribute an increasing level of trust to that information. In a scenario where data are incomplete and often different, this dynamic system is able to converge this information and assign quality levels to each attribute about an individual.”
Since there is no “ultimate truth” provided by entities like State authorities, it is only natural that there are slightly different views of what the attributes associated with citizens are. With the help of mathematics, it is possible to analyse a collection of data (statements) about citizens and confirm if they are reliable. Another innovative aspect of IDINA is the approach to individuals’ privacy. “The concern about data protection was central from the beginning. The system does not store sensitive information centrally, and data control is always in the hands of the citizens themselves,” reinforced João Marco.
“In an ideal world, IDINA would not be necessary,” recalled João Marco. But if this is not the case, it may be the answer not only for populations without civil registration services, but also for other contexts, like forced migration or marginalised groups. “IDINA can be a transitional solution for a state system or, alternatively, it can act as an independent system that ensures identity and access to vital services,” explained João Marco.
We are not a number
Working with digital identity is not new for Vítor Fonte. He was part of Mobile ID, a project that began with the Imprensa Nacional Casa da Moeda, based on the architecture and mechanisms proposed in ISO/IEC 18013-5:2021 for the construction of security documents.
“The idea was to use an architecture, considering a set of protocols, to create digital security documents, with different fields of application. It could be, for example, a driver’s license, a car registration document or even a football club membership document,” he stated. In essence, the work carried out was already meeting what is now the European Union’s goal of allowing citizens to associate their national digital identities with other personal attributes, like driving licences, diplomas and bank accounts, while providing full control over their data when accessing online services throughout the European space, eliminating the unnecessary sharing of personal information.
We now return to “Severance”: do we have the right to separate our identities as well? After all, we all play different roles in society and the multiplicity of identities would reduce the risk of traceability. Are we misidentified in current systems? “Basically, we can say that they are very partial systems. It’s a well-known limitation of traditional identification systems. They are used to identify us within a set of entities, but nothing more than that; there is a set of information associated with a person that gets lost – for better or for worse,” said Vítor Fonte.
The existence of a digital ecosystem with entities that produce information about citizens, and that can be used in different contexts, may well be the future. IDINA has already taken the first steps. “People are not numbers; in IDINA, when we talk about identity, we do not refer to the civil registration number. We focus on the interactions that define them, and we extract their identity from these elements.”
Now, picture this: how much easier it would be to make decisions if we knew people beyond what a system defines as proof of identity – whether it’s a fingerprint, a social security number, or a birth certificate. Does it seem impossible?
From theory to practice
“We have made a significant effort to have better information consolidation algorithms than what has been described in the literature so far. Therefore, there are truly innovative contributions in this area”, said Vítor Fonte. The next step is to take the solution from the lab to the field. “IDINA was lab-tested with public data and synthetic data generated by to validate the components of the system. Now we must test it in real-world scenarios, evaluate the result and make the necessary adjustments,” explained João Marco.
According to Vítor, IDINA be used in contexts in situations where there is not even a written representation of the name of the citizens: “when there is no written tradition, the same name can be written differently, and the IDINA must understand these contexts. Here’s an example: Vítor can be written as ‘Vítor’ or ‘Víctor’, so the system must consider this detail.”
There are also data and attributes that may not be thought out and that must be consolidated, as well as the issue of interoperability. How will IDINA connect to other systems? How do we motivate entities to participate? It is entirely possible that the next step is to migrate the IDINA data to a Central Civil Registry Service. “IDINA is a non-authoritative system, but it can serve as a basis for incorporating this information into an authoritative system, with legal bases that could support the use of this information,” said João Marco.
Deep down, IDINA represents a significant step at the intersection of technology, human rights and social inclusion. “Transferring technology from our laboratories to improve people’s quality of life is our real commitment,” concluded Vítor Fonte.
We return to the question from the beginning: if you are asked to identify yourself, how will you do it?
News, current topics, curiosities and so much more about INESC TEC and its community!
