The AI border guard: How algorithms are changing East Africa’s travel checks
A traveller arriving at an East African airport may see little more than a passport officer, a scanner and a queue. But by the time that traveller reaches the immigration desk, an automated system may already have processed information about the journey and assessed potential risks. Kenya provides one documented example of how artificial intelligence […]
When East African travellers arrive at an airport in the region, they may interact with a handful of personnel and a few machines. However, unbeknownst to many, an automated system may have already scrutinised the traveller's journey and determined potential risks before they even reach the immigration desk. A prime example of this emerging trend can be found in Kenya, where the integration of artificial intelligence into border management has gained traction in recent years.
In 2023, Kenya adopted the Electronic Travel Authorisation (ETA) system as a replacement for its previous e-Visa system. To bolster border security, Kenya enlisted the services of Swiss company Travizory, which provided a cutting-edge border security system. Travizory's solution incorporated real-time carrier data, AI-powered profiling and risk-assessment, as well as automated alerts.
The system's proponents claim that it facilitates proactive intervention and informed decision-making to prevent potential threats from breaching the border.
Following the implementation of the Travizory system in Kenya, the same technology continued to operate in Seychelles, where it had been in service since 2021. However, the report does not offer definitive evidence that the Travizory system is currently deployed across all East African nations, including Kenya, Uganda, Tanzania, Rwanda, Burundi, South Sudan, the Democratic Republic of Congo, and Somalia.
Instead, the report serves as a starting point for a broader inquiry into the implications of AI-driven border management in the region.
The report raises several concerns regarding the use of AI in border and migration settings. One of the primary issues is the "black box" nature of many AI systems, which makes it challenging to ascertain exactly how they operate, what data they utilize, and what assumptions or biases they incorporate. As a result, travellers may supply their passport, visa information, or other travel details to immigration authorities, only to be left uncertain about how that information is processed, the patterns sought, and the circumstances that might trigger an automated alert.
The lack of transparency surrounding AI systems raises significant accountability concerns. It becomes difficult for civil society organisations and journalists to hold tech companies and governments accountable for potential ethical issues, as well as to influence the development of regulations governing AI in border management.
The report indicates that 45% of surveyed civil society organisations cited a lack of transparency from those deploying AI about when and how systems are used as a hindrance to oversight. Similarly, 38% of respondents identified a lack of transparency from developers regarding the inner workings of these systems as a barrier to understanding and regulation.
Another considerable concern pertains to the regulation of AI in Africa. One respondent noted that there are "weak and sometimes non-existent regulatory frameworks in Kenya and much of Africa," suggesting a lack of comprehensive AI governance laws and enforcement mechanisms. Additionally, existing data protection laws, such as Kenya's Data Protection Act, may not be adequately enforced or adapted to the unique challenges posed by AI.
The implications of these concerns extend beyond civil society. Journalists, researchers, activists, business travellers, tourists, migrants, and ordinary citizens all encounter systems designed to facilitate rapid security decisions. The central question for the public becomes whether a machine is increasingly influencing decisions without the individuals being assessed being aware of it.
Kenya's experience with Travizory offers a glimpse into the hidden layer of border security, where the most consequential decision-maker may no longer be the official standing behind the desk but rather the system the traveller never sees. As reporters across East Africa aim to uncover the specific systems being utilized, the information processed by these systems, the companies supplying them, the oversight mechanisms in place, and the rights of travellers when an algorithm raises an alert, the focus remains on unveiling the extent to which AI is shaping the future of border management in the region.
Written by urgent.news from People Daily Kenya's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
