Opinion
Is AI stealing our souls? The ethical questions behind Gulf’s AI boom
In Abu Dhabi, crews are pouring concrete for a five-gigawatt AI campus, the largest outside the United States. Saudi Arabia’s national champion, HUMAIN, is racing to build gigawatts of compute and has launched an initiative (SAMAI) to train a million citizens in AI use; the kingdom now ranks first in the world for government AI ambition on the Tortoise Global AI Index.
QIA-backed QAI and Brookfield have launched a $20bn AI infrastructure venture, while the UAE’s MGX investment platform is investing across the AI stack, from semiconductors to applications.
By 2025, 84% of organisations across the GCC reported using AI in at least one business function, up from 64 percent two years earlier. National strategies project substantial gains in non-oil economic output from AI over the coming decade.
The Gulf is not alone, investment in AI is almost ubiquitous, as countries fear being left behind in the AI revolution. Fears that tools like Claude, ChatGPT, Midjourney will lead to job losses have failed to materialise to the extent feared. However, the much vaunted productivity gains have also failed to materialise.
Yet other, perhaps more important concerns remain, such as the impact of AI on humanity, and crucially, our brains, and the essence of what makes us human — the soul. By "soul” I mean the spiritual breath of life placed within humans. For some that may be the divine or immortal entity that exists beyond the material world. For most it will be the entity that possesses consciousness, free will, and a moral capacity.
At a recent event held in Education City by WISE, entitled Trust Me, I’m an Algorithm?, we discussed the impact of AI on higher ed. Specifically, we talked about the need to harness AI and avoid the type of cognitive offloading that AI is allowing. (Cognitive offloading is where we ‘offload’ our thinking to machines, theoretically sacrificing a learning opportunity).
Yet the longer I spend in higher education the more despondent I become. I worry what dependence on AI is doing to our ability to think independently and critically. When I teach students I use a useful metaphor about the jewel wasp put forward by Matt Simons in *Plight of the Living Dead: What Real-Life Zombies Reveal about Our World — and Ourselves.
The jewel wasp hunts cockroaches several times its size, but it does not kill them. It stings twice, once to paralyse the front legs, then precisely into the brain, depositing venom in two spots that govern the roach’s drive to move and flee. The active ingredient includes dopamine. Dopamine is the same neurotransmitter the human brain releases at a social media notification. It’s often referred to as the feel good chemical.
The roach is not injured; it can walk, right itself, and run if it chooses. What it loses is the impulse to use those intact abilities on its own behalf. It is, by every sign, content, grooming itself calmly as if nothing is wrong. But it has effectively become a zombie. When the wasp tugs the stump of an antenna, the roach follows it into a burrow where the wasp’s larva will eat it alive, and never resists. The roach becomes a living host, placated through dopamine, directed by an external entity that uses its body as a host.
The fairly obvious comparison is that social media, and now AI, are the things that are slowly supplanting our ability to think independently and critically, in many cases feeding us dopamine as we celebrate the technologies that may be responsible for the decline of our ability to act independently. We may be alive, but we are slowly dying, succumbing to the power of external technologies.
The empirical work describes the negative impacts of AI. In a 2026 study by researchers at Carnegie Mellon, MIT, Oxford, and UCLA, participants solved ordinary problems, fractions, reading comprehension, some with an AI assistant that could solve them outright. When it was removed, those participants were far more likely to give up or answer wrong, an effect that appeared after about ten minutes. As co-author Michiel Bakker of MIT frames it, persistence, the willingness to stay with a hard problem, is what lets a person learn at all. It is the first faculty to atrophy, and the one the others depend on.
A skill that goes unused inevitably decays. A separate study from MIT’s Media Lab recorded the brain activity of people writing essays: those who let a language model write showed lower engagement and weaker memory signatures, and afterwards could not reliably explain what they had submitted. They produced competent work and retained nothing from producing it. What are we as humans if we cannot remember the work we produce? Writing is thinking, and thinking is a key element of what defines our humanity.
The common argument is that we should ban the tools. The issue is how to frame the choice as one between AI that replaces human reasoning and intelligence amplification (IA) that enhances it, between a tool that returns the answer and one that questions, and sometimes withholds it so the user has to think. The first is cheaper, easier, and requires us to use less of our brain. The second is the only one that leaves the user more capable. The thing is, it is easier to say this than realise it in practise, but these are key questions that we need to grapple with.
The Gulf’s AI race is usually framed in terms of economic diversification, geopolitical influence, and technological leadership. But beneath the language of infrastructure and productivity lies a deeper question about what these systems are doing to us as human beings.
The danger is not that AI becomes human, but that humans become less so; less thoughtful, less reflective, less capable of independent thought. Such elements are central to creativity, morality, and learning itself. They are central to our identity too, whether as humans, or what makes us uniquely Khaliji.
This does not mean rejecting AI. It means recognising that convenience is not always progress, and that intelligence amplification must take precedence over intellectual substitution. Training people in AI also means drawing attention to the deeper questions about the human condition and AI. The question for the Gulf is therefore not simply whether it can be a leader in the AI revolution, but what steps it is taking to prevent eroding the very human capacities it hopes to cultivate.
- The writer is an Associate Professor of Media Nalaytics at Northwestern University in Qatar, a Unesco Fellow, and author of several books including Digital Authoritarianism in the Middle East.