Eleanor Drage's new book, What If We Got AI Right?, exposes how artificial intelligence algorithms perpetuate and entrench pre-existing biases, a critical insight for understanding Europe's increasingly automated border regime. Drage, an assistant research professor at the University of Cambridge, warns that the dangers of AI in law enforcement are profound. These systems, built on hours of human labour, are not neutral tools; they reflect the discriminatory structures of the societies that create them.
AI's Racist Logic in Law Enforcement
The book argues that tech giants, chasing profit and power, have become too big to care about ethics. This unchecked power allows AI to be rolled out without consent and without consideration for planetary health. For those targeted by Europe's border apparatus, this means surveillance systems and predictive policing tools, often powered by AI, are designed to criminalise movement and reinforce racial profiling. The "migration crisis" is a manufactured moral panic, and AI provides new tools for its enforcement.
Drage's research specifically highlights the dangers of AI in law enforcement. When algorithms are inherently biased, their application in policing and border control inevitably leads to disproportionate targeting and unjust outcomes for migrant communities. This structural racism isn't a malfunction of the system; it's often a feature, designed to deter and exclude.
The author contends that a cultural obsession with AI-induced apocalypse distracts from practical conversations about making AI safer. Instead, citizens need more power over how their information is used and how AI models are trained and regulated. This call for democratic control directly challenges the opaque operations of the border regime, where biometric databases and surveillance technologies are deployed with little accountability.
The Invisible Labour of AI
Drage reveals that AI's supposed "humanity" stems from countless hours of human labour. This includes workers in silicon and quartz mines, microchip factories, and data labelling centres. This globalised, often exploited labour force is the true engine behind the "intelligence" sold by big tech. The illusion that the labour force is disappearing is just that – an illusion; work is simply shifting, from artists to data labellers and from marketeers to engineers. This mirrors the exploitation of migrant labour, where capital moves freely across borders while workers are criminalised for doing the same.
The "cloud," Drage reminds us, is simply someone else's computer. An AI "hallucination" is a system error or a data labelling mistake. These seemingly technical issues expose the human fallibility and the inherent biases embedded by those who design and train these systems. The pursuit of profit by a narrow few executives drives the development of AI, not a commitment to universal human rights or equality.
Challenging the Tech Border Regime
Drage co-designed a toolkit to help AI companies act ethically and comply with EU regulations, describing it as a "cornerstone of utopianism." This shifts the focus to how AI is built and made, rather than its abstract potential. Utopia, she states, is not a destination but the "process of working together to develop a fairer world." This perspective aligns with migrant solidarity networks and anti-deportation campaigns, which actively work to dismantle the structures of Fortress Europe.
Examples like the Mumkin app, designed to facilitate difficult conversations about female genital mutilation in India, and Kuini, an AI chatbot built to help Māori women quit smoking, illustrate AI's potential when developed with community consent and for liberation. These projects stand in stark contrast to the deployment of AI for border enforcement, which serves to divide and control. The EU's new Migration Pact, outsourcing asylum screening to unstable third countries, relies heavily on technological infrastructure that will inevitably incorporate biased AI, further entrenching the racist double standard in Europe's asylum system.
The question posed by the book, "What if we got AI right?", remains unanswered by the reviewer. But for those fighting for open borders and human equality, getting AI right means dismantling its use in the border regime and ensuring its development serves collective liberation, not corporate profit or state control.