Depending on who you ask, India’s national biometric ID, Aadhaar, is either a massive success or a case study on how not to create a state ID system.

Among its fans: government officials, technology corporations who benefit from its implementation, the World Bank.

Among its detractors: technical security experts, advocates of the poor and minorities, public interest lawyers, citizens concerned about transparency, rights, liberty and security.

The promise of Aadhaar, which now logs the fingerprints, iris scans (and soon, facial recognition) of 1.19 billion citizens in India, was to eliminate duplicate records and fraud in the distribution of state welfare. Biometric authentication was sold as a foolproof technology. Armed with a unique ID, the poorest would have easier access to bank accounts, SIM cards and more.

And it would be voluntary.

Around the world – not just in India – more than a billion people are never officially registered or counted at birth. This absence of demographic data and statistics of the poorest regions complicates the creation of public policy and the delivery of aid. Digital ID can help solve this, but there are multiple ways to design such systems, and not necessarily with biometrics.

The problem with Aadhaar is a tendency to disregard or underestimate the serious consequences for those for whom the system does not work. Let’s call it the willful ignorance of officials and vendors to acknowledge the lived experience of those who interact with the system and its reported flaws. India is aggressively pushing forward with linking the Aadhaar number to a variety of government and private databases. Too hastily? India has not yet enacted a comprehensive data protection law that upholds the right to privacy.

As long as officials insist on the infallibility of biometric identification (as well as on the security of flawed custom APIs and protocols) the countless people who experience “error” messages because of faded fingerprints or faulty data entry have limited opportunity for corrections or to seek alternatives. Aadhaar is allegedly voluntary, but you need it to get married, buy property, get a bank account or even… track a lost Amazon package. The 12-digit identification number is requested so often, only very few holdouts resist.

When vast distances in rural India must be travelled to visit a bank (and electricity or Internet services regularly malfunction and require you to return another day), low tech alternatives might not only be preferable, but mean life or death. The law says these analog alternatives must exist, but the reality is that this usually isn’t the case on the ground.

It is a vulnerability that sensitive biometric data is stored across databases that are centrally linked – especially since the project has yet to receive an independent security audit. Aadhaar data of more than 1 billion people can be found for just $8 USD on the black market, and if your profile is misappropriated, you can’t ever exchange your biometric data for a new set.

A feature of Aadhaar is that both the private and public sector can incorporate it in their products and transactions. This audit trail of Aadhaar authentications across public and private spheres are readily available to the government with court warrants. Vocal concerns in India about Aadhaar as the technical foundation for a surveillance state are not overblown.

The Supreme Court of India has given Aadhaar a look over more than once and will likely do so again thanks to the privacy and welfare activists of India who are pushing back aggressively against the most egregious problems with Aadhaar, and encouraging others to do the same around the world where similar initiatives are in the works. You should too.

Further reading:

Aadhaar debate: Privacy is not an elitist concern – it’s the only way to secure equality, Malavika Jayaram, 2015
Rethink Aadhaar
Aadhaar or else, Jean Dreze, 2017
Identity Policies: The clash between democracy and biometrics, Privacy International, 2017