Aadhaar Conundrum: Three Tests for a Fool-Proof Identity

Is the Aadhaar system as a method of unique identification better than what we had before? Will it curb corruption and fraud even if India’s bureaucratic and administrative apparatus doesn’t play ball?

How safe is the data collected for Aadhaar cards? Representative image. Credit: Reuters/Mansi Thapliyal

How safe is the data collected for Aadhaar cards? Representative image. Credit: Reuters/Mansi Thapliyal

The debate over India’s biometric authentication programme, Aadhaar, has only intensified after the recent Supreme Court order on privacy.

Started in 2009, the digital identification project has continuously expanded in scope, despite restraining Supreme Court orders. Today, it covers almost every conceivable facet of a normal civic life. While this has fanned the fears of an Orwellian state subsuming the inalienable sovereignty of individual citizens, there is a crucial need to revisit the basic problem statement that led to its genesis.

Aadhaar was conceived to solve the problem of fake and ghost identities. All the present uses of Aadhaar profess to leverage this purported capability of it. Thus, we have Aadhaar getting linked to ration cards, NREGA payments, PAN cards, bank accounts, mobile SIMs, mid-day meals in schools, school admissions, university admissions…the list is endless. In a country plagued by corruption and bureaucratic inefficiency, the hope is that Aadhaar will help eliminate fraudulent identities, thereby saving the direct and indirect losses to the nation.

To correctly understand how far Aadhaar can go towards achieving its goals, it may be pertinent to list out the key features that we should look for in an “identity proof” and then see how Aadhaar stacks up against each of these.

Principle 1: It should be difficult to create fake identities

This particular vulnerability hampers nearly all identity systems, which therefore became the justification for introducing Aadhaar. In the case of Aadhaar, the shield of invincibility was built on the promise of biometric technology, i.e. biometric deduplication would instantly throw out anyone trying to enrol more than once. This indeed is an undeniable truth (assuming we discount a not-insignificant portion of the population that have obscure biometrics). During the PAN-Aadhaar hearings, the then Attorney General Mukul Rohatgi made no bones about stating that the 115 crore strong Aadhaar database has no duplicates at all.

Also read: How Mukul Rohatgi Set the Cat Among Pigeons on Privacy, and Now Says He Didn’t Intend To

The hidden script here is that a unique biometric profile is not necessarily a genuine one. How about a person interchanging his left and right hands when scanning fingerprints, or scanning the ten fingers in various different permutations, or pooling in biometrics from multiple persons? Each of these will also emerge as a unique biometric profile and pass the barrier of deduplication.

As if reading from the same script, a racket that had successfully created thousands of fake Aadhaar IDs, against bogus biometrics, was recently exposed in UP.  It thus appears that the promise of Aadhaar – elimination of fake and ghost identities – is nothing more than an appeal to the honesty of concerned (or empowered) individuals, which is no improvement from the situation without Aadhaar.

The involvement of private parties in Aadhaar enrolment has further magnified the possibility of fraud to such an extent that law enforcement is likely to be overwhelmed.

In all fairness, it seems the original designers of Aadhaar had anticipated the creation of fake profiles, which is why it was always meant to be a paperless ID, to be validated only through biometric authentication. But biometric authentication is a beast in its own right, afflicted by a range of variable factors that are beyond human control. So the guard has been selectively lowered, depending upon the target population. While biometric authentication is being enforced with a vengeance in PDS disbursals and NREGA payments, it has been neatly given the slip for the PAN-Aadhaar linkage.

The PAN-Aadhaar linkage should, therefore, be considered an exercise in nullity. All that is needed to validate a fake PAN is to ‘generate’ a fake Aadhaar with matching demographic details.

Biometric authentication for Aadhaar. Credit: Reuters

Biometric authentication for Aadhaar. Credit: Reuters

That the Unique Identification Authority of India (UIDAI) has given up on biometric authentication in certain cases is clear from the fact that Aadhaar is now allowed to be authenticated through mobile OTP and in some quarters even acceptable as a physical “Aadhaar card”. These improvisations are only to keep Aadhaar ticking for the law-abiding citizens, while they are powerless to stop the usage of fake Aadhaar IDs. A fake Aadhaar ID can jolly well be used in perpetuity, validated through mobile OTP or as a physical “card”. In fact, the prospect of using Aadhaar without biometric authentication has opened up a lucrative market for fake Aadhaar IDs.

The curious fact is that while Aadhaar does struggle to counter fake identities, the original myth about its invincibility has seeped deep within the officialdom. That every government department is looking up to Aadhaar as the one-stop solution to eliminate leakage and fraud is an absurdity, to say the least.

Principle 2: A genuine person should not face difficulties in proving his identity

This is the most basic of all requirements. Of what worth is an identity proof if it cannot guarantee identity to its holder? The founding fathers of Aadhaar have placed the identity of all Indians on the thin ice of biometrics. Biometrics is not an exact science. It is suitable for forensic and surveillance purposes, where it is applied on a best-effort basis. But it would be perilous to leave the identity of an individual to the mercy of circumstances. Aadhaar identity is valid only if the individual’s current snapshot of fingerprints/irises matches the records in the database. It may trip any day, due to uncontrollable reasons like ageing, manual labour (that causes fingerprints to fade), illnesses (like cataract), etc. Above all, an overwhelming section of our population does not have biometrics fit enough for the kind of rigour Aadhaar demands. This has already dispossessed a not-insignificant number of people of their legitimate welfare claims.

It is not that only the poorer sections of our country need to dread biometric technology. Even the better-off sections are facing the reality of biometric authentication when attempting e-KYC for various purposes. The Aadhaar Act has provided an elaborate cover to the UIDAI, by making it the responsibility of the concerned individual to maintain his biometric records in the database. This is an unconscionable condition, as changes to biometrics are not cognisable events that the individual can comprehend (unlike say a change of demographic details).

Also read: Four Reasons You Should Worry About Aadhaar’s Use of Biometrics

Added to the above matrix is the dependency on power and network availability (for performing the authentication transaction). As mentioned in the earlier section, the UIDAI has suggested palliatives like mobile OTP, to tide over the vagaries of biometric authentication. Mobile OTP as a sole and sufficient authentication mechanism is too weak a security layer and should raise the hackles of genuine, law-abiding citizens.

The asymmetric power equation between the UIDAI and the Aadhaar number holders is explained in detail in another article.

Principle 3: It should be difficult to misappropriate the identity

The strength of an identity system is determined by how difficult it is to misappropriate someone else’s identity. The key to unlocking Aadhaar identity is biometrics, which is neither secret nor changeable, hence not the least suited to serve as a password. It is trivial to discuss here the methods by which fingerprints, iris images or other biometrics can be copied; sufficient to say that it takes only slightly more effort than a photocopy.

To understand the full import of this, let us consider the example of the BHIM-Aadhaar Pay app. The CEO UIDAI in a recent article sought to downplay the risks, saying that even other payment systems, like cheques, ATM cards, internet banking, etc. are susceptible to fraud. But is that a fair comparison? To break into internet banking, there is a need to hack the password and also intercept the second layer authentication. To make use of a forged signature, there is a need to either steal or replicate the bank cheque leaves. In all these cases, the aspect of security comes from that element in the transaction which is in the private possession of the individual. In the case of BHIM-Aadhaar Pay, all that is needed to complete the transaction are the Aadhaar number and the fingerprint, both of which are revealed at the merchant site. So the CEO UIDAI’s assertions can be likened to saying that you shouldn’t bother about locking your safe at all, simply because no lock can assure 100% safety.

Now extend this further to the full scope of Aadhaar. The combination of Aadhaar number and biometrics is considered as a sufficient token for opening a bank account, taking a mobile SIM and for many other purposes. That policy administrators have begun to consider Aadhaar as the “single source of truth” is a big reason to worry for honest, law-abiding citizens.

The talk about encryption of biometrics at the endpoints is nothing but a smokescreen because biometrics can be lifted by completely external means. It is also irrelevant to talk about penal consequences of such fraud. If the fear of law was sufficient to prevent fraud, Aadhaar would have never come into the picture. What’s worse is that once a user’s biometrics have been compromised, the Aadhaar identity becomes unusable for life.

Is the Aadhaar system, as a method of unique identification, better than what we had before? Will it curb corruption and fraud even if India’s bureaucratic and administrative apparatus doesn’t play ball? While that appears to be an unlikely outcome, the collateral damages are many and far-reaching.

L. Viswanath is engineering professional working in Bengaluru. He blogs at bulletman.wordpress.com.