Big Tech's Grip on Migration Policy

The growing reliance on artificial intelligence is transforming operations in healthcare, commerce, and even policymaking. Indeed, AI’s ever-expanding role in global migration policy is fuelling a new “AI arms race” and, in turn, tightening the relationship between the private sector and international governance. For some, this fosters necessary efficiency and innovation. For others, Big Tech sounds warily like “Big Brother,” skewing the ethics of consent and intent. Considering these implications - both the risks and benefits, to what extent should we be concerned by AI’s use in global migration policy?  

Supporters point to three pay-offs: increased efficiency, improved preparedness, and greater securitization of the border. AI has streamlined administrative tasks like processing travel documents, identity cards, and work permits. This includes biometric gates facilitating a “seamless experience for passengers” at major airports, as well as even automating asylum and visa applications. Growing databases also track and predict population movements and migrant integration outcomes, enabling countries to prepare accordingly. In 2018, researchers found that the use of machine learning to algorithmically place refugees increased their employment outcomes by 40 -70% relative to current assignment practices. Despite these supposed benefits, the allure of AI’s capabilities quickly deteriorates once ethics and intentions come into question. 

Considering consent and intent behind AI turns the possibilities of technological advancement into dystopian nightmares for critics, especially in regards to the treatment of refugees and asylum seekers. While the secrecy and lack of consent surrounding AI affects everyone, refugees and asylum seekers experience this ambiguity to the extreme. Due to their particular vulnerability, they are unlikely to object and have limited knowledge about how their personal information is being used. Especially in the involvement of third-parties, priorities are usually placed on efficiency and profit maximization, rather than their safety and security. 

Experimentation and digital tracking techniques can take on particularly sinister connotations. Even if well-intentioned, there is a degree of distasteful violation in the UN’s 2017 Unite Ideas Internal Displacement Event Tagging and Extraction Clustering Tool ‘challenge’ which sought a ‘winning’ team to track and analyze refugees through machine learning. Human rights critiques have disparaged ‘competitions’ and studies such as these for using refugees as experimental subjects to train AI algorithms. 

Additionally, the UN’s use of fingerprint and iris scans to administer food aid in the Middle East is disconcertingly intended to determine who should and should not be fed amongst a vulnerable population. They have also proven to be fallible with an error rate of 2-3%, suggesting that 11,800 claimants have been wrongfully denied aid. Possible breaches of sensitive information and databases also risk further discrimination, involuntary repatriation, and persecution of refugees. 

Although these issues appear far away, the ever-increasing reliance between Big Tech and global governance makes them far more pressing than we realise. DNA services including Ancestry.com have been used by governments to confirm the nationality of individuals considered for deportation. Even humanitarian organizations, such as the World Food Programme, involve themselves with ethically questionable partners to provide analytical services. Ironically, for the WFP, Palantir uses the same criticized techniques they used with the United States Bureau of Immigration and Customs Enforcement to reveal the location of migrants and facilitate their expulsion.

While these critiques barely scratch the surface of AI’s grip on global migration policy, they make it difficult to view AI without qualification. Indeed, both sides illuminate how AI is increasingly connecting the world. For better or worse, AI’s involvement in migration management will clearly only continue to grow. Therefore, placing ethics at the forefront is essential in keeping its eventual use in check. While there is certainly room for concern, I remain hopeful that AI has potential for good, so long as it is placed in the right hands, within an ethical framework considerate of those particularly vulnerable. With the way it is going, it appears we have much work in reaching this ideal.

Sources

Bansak, Kirk, Jeremy Ferwerda, Jens Hainmueller, Andrea Dillon, Dominik Hangartner, Duncan 

Lawrence, and Jeremy Weinstein. 2018. “Improving Refugee Integration through 

Data-Driven Algorithmic Assignment.” Science 359 (6373): 325–29. Doi:10.11

26/science.aao4408. 

Bircan, Tuba, and Emre Eren Korkmaz. 2021. “Big Data for Whose Sake? Governing Migration 

through Artificial Intelligence.” Humanities and Social Sciences Communications 8 (1). 

doi:10.1057/s41599-021-00910-x. 

Fournier-Tombs, Eleonore. 2021. “Towards a United Nations Internal Regulation for Artificial 

Intelligence.” Big Data & Society 8 (2): 205395172110394. doi:10.1177/

20539517211039493. 

Khandaker, Tamara. 2018. “Canada Is Using Ancestry DNA Websites to Help It Deport 

People.” VICE. VICE. July 26. https://www.vice.com/en/article/wjkxmy/canada-is

-using-ancestry-dna-websites-to-help-it-deport-people. 

Moore, James. 2020. “Heathrow Installs Biometric Gates to Improve Customer Experience 

While Maintaining Security Standards: IFSEC Insider.” IFSEC Insider | Security 

and Fire News and Resources. IFSEC Insider. October 1. https://www.ifsecglobal.

com/access-control/heathrow-installs-biometric-gates-to-improve-customer-exper

ience-while-maintaining-security-standards/. 

Madianou, Mirca. 2019. “The Biometric Assemblage: Surveillance, Experimentation, Profit, 

and the Measuring of Refugee Bodies.” Television & New Media 20 (6): 581–99. 

doi:10.1177/1527476419857682. 

Nalbandian, Lucia. 2022. “An Eye for an ‘I:’ A Critical Assessment of Artificial Intelligence 

Tools in Migration and Asylum Management.” Comparative Migration Studies 10 (1). 

doi:10.1186/s40878-022-00305-0. 

Raftree, Linda, and Karl Steinbacker. 2019. “Head to Head: Biometrics and Aid.” The New 

Humanitarian. The New Humanitarian. November 24. https://www.thenewhuman

itarian.org/opinion/2019/07/17/head-head-biometrics-and-aid. 

United Nations AI Advisory Body. Interim Report: Governing AI for Humanity. United Nations, 2023.

Image source: https://www.independent.co.uk/news/business/business-reporter/ai-enhance-future-digital-transformation-b1871195.html