BIAS INBUILT

Exploitative use of AI by governments and private entities threatens humanity, emphasising the need for strong guardrails

- 

IMAGINE LIVING in an active war zone, battling for daily survival. What you may not know is that you are also being "scored"-on the basis of who you know socially, who you chat with online, and how often you change your phone or address. And your score decides if you "can be marked as a target subject to attack." This may be a reality in the Gaza Strip, says Human Rights Watch. In September 2024, the research and advocacy organisation published an article on Israel's use of digital tools against Palestine. Among these is "Lavender", an artificial intelligence (AI) tool that probably assigns Gaza residents "a numerical score relating to the suspected likelihood that a person is a member of an armed group". If scores exceed a threshold set by Israeli military, the individual may be targeted, it says.

As AI is taking over all fields, defence and warfare are not left out. And it is not used for just surveillance or strategy. Countries like China, Israel, Russia, South Korea, Türkiye, UK and the US are investing in building autonomous weapon systems (AWS), removing the need for ground troops, says a 2024 article by the Carnegie Endowment for International Peace, a think tank headquartered in the US. The US Department of Defense defines AWS as systems that "once activated, can select and engage targets without further intervention from a human operator." This raises concerns about the potential of the AI-based weapons going rogue or humans deflecting responsibility. The algorithms could also be built from biased and incomplete data. For instance, Israeli military designated Palestinian human rights organisations as "terrorist groups".

If such broad definitions were used in training of tools like Lavender, it could increase the possibility of civilians being targeted, says Human Rights Watch.