An analysis of the Biometrics in the Unique Identification project
The ICTs (Information and Communication Technologies) have always been seen as an effective tool of governance across the world. Since it is an idea which came up with modernity, the modern state has used it widely in governance and administration, also sometimes with immense interference of market forces.
There cannot be any other recent, well suited example of ICT initiative in India other than the ongoing UID (Unique Identification) project. It is a project which sets to uniquely identify every resident of the country by providing a unique identification number. And at the same time it also envisages to maintain a centralized database of individualâs demographic and biometric information. Â The argument behind the collection of biometrics is its requirement in the de-duplication process, henceforth assuring that the data stored in the central database i.e. CIDR (Central Identities Data Repository) is clean and unique. In this context, it must be clear that the biometrics is referred to as the facial image, finger prints and iris scans.
While stating its objectives the UIDAI (Unique Identification Authority of India) very clear says that âit envisions full enrollment of residents with a focus on enrolling Indiaâs poor and under-privilegedâ with âthe UID method of authenticationâ, that âwill improve service delivery for the poorâ.[1] It has been claimed by the UIDAI authorities and is also evident in the UIDAI documents that through better targeting with the help of the UID the flagship social welfare schemes of the government including NREGS, PDS and RSBY will see drastic improvements.
Is Biometrics a reliable technology?
The UIDAI largely relies on the biometrics for the authentication of identity. It becomes significant in the supposition that any kind of technological error or failure specifically relating to biometrics could adversely affect the UID linked services importantly NREGS or PDS. Several cases of biometric failure have been argued, as of now, by scholars and biometric experts.
R.Ramakumar in his paper, âthe unique Id project in India: A skeptical noteâ discusses two debated and controversial cases of false matching of finger prints in the USA that led to massive protests from the human rights activists :
   â âŠthe case of Brandon Mayfield, an Oregon attorney and Muslim convert who was held for two weeks as a material witness in the Madrid bombing of March 11, 2004âŠMayfield, who claimed not to have left the United States in ten years and did not have a passport, was implicated in this attack almost solely on the basis of a latent fingerprint found on a bag in Madrid containing detonators and explosives in the aftermath of the bombing. Unable to identify the source of the print, the Spanish National Police emailed it to other police agencies. Federal Bureau of Investigation (FBI) Senior Fingerprint Examiner Terry Green identified Mayfield as the source of the latent print. Mayfieldâs print was in the database because of a 1984 arrest for burglary and because of his military service. The governmentâs affidavit stated that Green âconsiders the match to be a 100% identificationâ of MayfieldâŠ
 âŠA few weeks later the FBI retracted the identification altogether and issued a rare apology to Mayfield. The Spanish National Police had attributed the latent print to Ouhnane Daoud, an Algerian national living in SpainâŠ
 âŠBut the Mayfield case was not the first high-profile fingerprint misattribution to be exposed in 2004. In January, Stephan Cowans was freed after serving six and a half years of a 30- to 45-year sentence for shooting and wounding a police officer. Cowans had been convicted solely on fingerprint and eyewitness evidence, but post-conviction DNA testing showed that Cowans was not the perpetrator. The Boston Police Department then admitted that the fingerprint evidence was erroneous, making Cowans the first person to be convicted by fingerprint evidence and exonerated by DNA evidence.â[2]
 It is to be taken into consideration that the above two cases are from a developed country where ICTs have influenced public policy and subsequently, the functioning of the state. For instance, biometrics, especially finger print technology has so far been used in forensic sciences, mainly for criminal investigation. However, both the cases cited previously have pointed out the flaws regarding the accuracy of such technology and also consequently its reliability. In a similar manner, the lack of careful operational processes in such a high magnitude application of biometrics in UID could result into high rate of technological errors.
The Biometrics Standards Committee of the UIDAI in one of its report foresees some challenges on the biometrics accuracy. The report says, âAn accuracy rate (i.e., True Acceptance Rate) of 99% is reported in the test of commercial system performance. Two factors however raise uncertainty on the extent of accuracy achievable through fingerprints: First, the scaling of database size from fifty million to a billion has not been adequately analyzed. Second, the fingerprint quality, the most important variable for determining accuracy, has not been studied in depth in the Indian contextâ[3].
 Here, I should cite some field experiences from my research work which highlights the operational hindrances appearing in biometrics capturing at UID enrollment sites.
Being a highly technological driven process, the enrollments are encountering problems at the same rate. At Mongolpuri enrollment camp in Delhi, Prabhjeet Singh of the enrollment agency (Strategic Outsourcing, a Bengaluru based IT company) says that biometrics capturing is not going slickly. In cases of good quality finger prints and iris scans the ideal scale of above 50 percent shows a strong signal. But it does not happen with bad quality finger prints. Like, with people involved in rigorous manual work whose finger prints gets altered and children below five years of age whose finger prints are not well developed, the capturing becomes difficult. âIn such casesâ, he says, âWe have instructions to do force capture after four attempts. It is during the de-duplication and verification process that the best one out of the four is selected to store the data into the Central Database, CIDRâ.
This method of forcibly capturing biometrics could be troublesome in future, as it could deprive the poor to seek benefits from the UID linked services. Say, if a manual labour goes to the PDS shop to get subsidized grains and if his fingerprints mismatch the one in the stored data, he will not get the benefits, as the identity is not proved. Such cases have been argued previously in several critical arguments.
The UID enrollment proof-of-concepts report identifies four possibilities of errors in relation with biometrics. Â
FPIR: False Positive Identification Rate: This is the likelihood that a personâs biometrics is seen as a duplicate (i.e., the biometric de-duplication software identifies his biometrics as matching with that of a different person), even though it is not a duplicate in reality.
 FNIR: False Negative Identification Rate: This is the likelihood that a person enrolls a second time and the de-duplication software is unable to identify their biometrics as a duplicate set.
FAR: False Accept Rate: This is the likelihood that a personâs biometrics is matched against a different person and the biometrics is seen to match, i.e. the person is wrongly seen to be a different person.
 FRR: False Reject Rate: This is the likelihood that a personâs biometrics does not match against an earlier sample of his or her biometrics and so he or she is not recognized as the same person.[4]
The case cited above, as per my understanding, could give rise to all the possible errors and particularly the last one.Â
Yet another finding from my research is again an erroneous element in the technological proceedings in biometrics. On conditions of not disclosing the name a technical advisor of the enrollment agency Strategic Outsourcing says, âsets of information (demographic and biometric) of individuals known as packets are needed to be sent for de-duplication. Every packet needs to be verified by a live or active operator through his thumb print. Ideally, the software should not accept any other thumb prints but it is happening. It is a major software fault and could be misused in many waysâ.
On this issue, the IT head of Strategic Outsourcing (Delhi region) N. Sudhakar says âit is impossible. It cannot happen. And if it is happeningâ he admitted âthe packets will be rejected during the de-duplication process because of not been verified by an authentic supervisorâ.
Not quite certain of the de-duplication accuracy, the above stated software fault could have two possible drawbacks. One, if such packets gets ahead of the de-duplication successfully the probabilities of fake identity creation goes up. Second, as per N. Sudhakar, if they get rejected, the person fails to get an UID.
Role of the market and threat to security
In a globalised and liberalized economy, where private corporate becomes key players in influencing public policy, the case of biometrics being monitored by the foreign companies cannot be disregarded. A news story in tehelka magazine howls similar concern;
âThe three foreign companies will be implementing biometric solutions for UIDAI that includes designing, supplying, installing, commissioning, maintaining and supporting the multi-modal automatic biometric identification subsystem.
...the involvement of companies such as American defence contractor L-1 Identity Solutions â which has names associated with the CIA and other US defence organisations in its top management â together with US-based Ernst & Young and Accenture raises queries about how much access they will have over Indian data.
Among the names associated with L-1 Identity Solutions are former CIA director George Tenet and former Homeland Security deputy secretary Adm James Loy, also on the board of Lockheed Martin. The companyâs links with the US military establishment is underscored by the fact that its board of directors include former Army Technology Science Board member BG Beck, former chairperson for the Secretary of the Armyâs National Science Centre Advisory Board Milton Cooper.
Itâs feared that the database can be used as a bulwark against India because all USbased firms are subject to the Patriot Act that obligates American companies to share their data with Washington.â[5]
 Moreover, on the applications of the UID, for instance in the NREGS, an observer of the UID project and NREGS activist Nikhil Dey( Mazdoor Kisan Shakti Sangathan) says âall that the biometrics system can ensure is that the person who is drawing the wage is indeed the person who is putting his thumb impression but it cannot ensure that he went to work or he did how much work and he got how much wage? And that is where fraud is taking place and any biometric system cannot eliminate such kind of corruptionâ.
The way the UIDAI is propagating the merits of a biometric system in addressing the hitches in socio-welfare schemes does not seems to be realistic and acceptable to the experts. And henceforth, this ICT initiative has been widely criticized by activists, practitioners and in academia on various grounds.
                      Â
[1] http://www.uidai.gov.in/UID_PDF/Front_Page_Articles/Documents/Strategy_Overveiw-001.pdf
[2] Ramakumar.R, The Unique ID Project in India: A Skeptical Note, 2010
[3] Biometrics Design Standards for UID Applications, Version 1.0, Biometrics Standards Committees Report, UIDAI, 2009
[4] UID Enrollment Proof of Concept Report, http://www.uidai.gov.in/images/FrontPageUpdates/uid_enrolment_poc_report.pdf
[5] http://www.tehelka.com/story_main49.asp?filename=Ne210511TROUBLE.asp