Image: Getty Images/ For illustrative purposes
Ninety-four per cent of data leaders in the UAE admit they do not have full visibility into how their AI systems make decisions, according to a new report from tech company Dataiku, highlighting rising governance risks as the country accelerates its AI ambitions.
The Global AI Confessions Report: Data Leaders Edition, conducted by The Harris Poll for Dataiku and based on a survey of more than 800 senior data executives across eight countries, shows the UAE’s rapid AI adoption is being undermined by traceability and explainability gaps.
Only 17 per cent of UAE respondents said they always require AI systems to “show their work”, one of the lowest levels globally, while nearly two-thirds, or 62 per cent, said they were not confident their organisation’s AI systems could pass a basic audit of their decisions.
Despite these concerns, 72 per cent said they would trust an AI agent to make autonomous decisions in critical business workflows even if it could not explain its outputs, suggesting that implementation is outpacing oversight.
Just half of UAE respondents said they had ever delayed or blocked an AI deployment due to explainability concerns.
The findings come as the UAE pushes ahead with its National Strategy for Artificial Intelligence 2031 and seeks to position itself as a global hub for AI innovation.
Report finds leaders face pressure when it comes to AI adoption
The report, however, shows data leaders feel mounting pressure from the top.
Nearly six in 10 respondents, or 59 per cent, said their C-suite overestimates the accuracy of AI systems, while 64 per cent believe leadership underestimates the time and complexity needed to make AI production-ready.
A third of UAE data leaders, 32 per cent, reported being asked to approve an AI initiative that made them uncomfortable, and 75 per cent said their company’s AI strategy is driven more by technological ambition than by business outcomes.
Global expectations around accountability also appear weaker in the UAE.
While 56 per cent of data leaders worldwide believe a CEO will be forced out by 2026 over an AI-related failure, only 35 per cent of UAE respondents share that view, the lowest among surveyed markets.
More than half, or 53 per cent, said they do not feel their own roles are at risk if their organisation fails to achieve measurable business gains from AI in the next one to two years, marking the highest perceived job security in the study.
“An alarming revelation of the report is that enterprises in the UAE, much like those globally, are betting on AI they don’t fully trust, and doing so with a sense of confidence that they won’t bear much consequence if things go wrong,” said Florian Douetteau, co-founder and CEO of Dataiku. “The encouraging news is that governance challenges, such as explainability and traceability, can be overcome.”
Despite the risks, the study also found UAE data leaders are prioritising accuracy above financial considerations, with 57 per cent naming performance reliability as their top priority compared with only 10 per cent who cited cost.
Local executives remain cautious about applying AI to sensitive functions: 55 per cent said they would never allow AI agents to make hiring or firing decisions, 48 per cent would exclude AI from legal or compliance work, and 39 per cent would avoid using AI for mental health or employee wellness support.
“UAE organisations are leading the charge in adopting AI at scale, but the findings show that work must still be done to ensure responsible growth remains the priority,” said Sid Bhatia, Area VP and GM – Middle East, Turkey & Africa at Dataiku.
The Harris Poll conducted the research online from August, surveying 812 data leaders across the US, UK, France, Germany, the UAE, Japan, South Korea and Singapore.
Respondents work for companies with annual revenue of at least one billion dollars or regional equivalents and hold titles ranging from vice-president to C-suite level.
Dataiku, backed by investors including Wellington Management, Battery, CapitalG, ICONIQ, and FirstMark, positions itself as The Universal AI Platform, designed to deliver explainable and scalable AI across enterprises.
