International SOS Warns AI Misinformation Risks Are Rising as Adoption Accelerates

As the UAE continues to accelerate the adoption of artificial intelligence (AI) across both the public and private sectors, new research from International SOS  reveals that organisations increasingly view AI as a valuable tool for verifying risk information. At the same time, businesses are becoming more concerned about the growing threat posed by misinformation, deepfakes and synthetic content.

The findings, published in the International SOS Risk Outlook 2026 Pulse Check, come as AI adoption in the UAE outpaces many global markets. According to the report, nearly 60% of risk professionals surveyed believe AI’s greatest value lies in verifying and validating information from multiple sources, highlighting its growing role in helping organisations manage complex risk environments.

Despite this optimism, the research underscores significant concerns about the reliability of AI-generated information. More than four in ten respondents (42%) said their organisation has either been directly affected by false, misleading or unverified information related to health, wellbeing or security risks, or lacks sufficient visibility to determine whether such information has had an impact. Respondents identified fake signals creating operational uncertainty, synthetic content damaging credibility, and AI-powered deception and impersonation as key challenges.

These risks are becoming increasingly relevant in the UAE, where cybersecurity experts have warned that AI-driven disinformation campaigns and deepfakes are emerging as major threats capable of eroding trust, disrupting operations and influencing decision-making across organisations.

Commenting on the findings, Cvete Koneska, Global Security Director at International SOS, said AI offers significant potential to rapidly surface, synthesise and prioritise information, but cautioned that verification remains a task requiring human expertise. She warned that relying too heavily on current AI capabilities could amplify misinformation risks and delay critical decisions affecting people’s health, safety and security, advocating instead for a human-led approach to integrating AI into risk management.

The report also highlights growing concerns about AI-generated health misinformation. Irene Lai, Global Medical Director at International SOS, warned that deepfakes featuring real healthcare professionals spreading false information during disease outbreaks represent a new form of biowarfare that can be launched digitally without the use of biological agents, posing serious risks to public health and emergency response efforts.

Beyond AI-related concerns, the research found that many organisations continue to struggle with broader risk governance and decision-making. Only 10% of respondents said their organisations are able to respond “very quickly” when new risks emerge, while 68% described their response as “fairly quick.” However, one in five organisations (20%) admitted they are not quick to respond at all, raising concerns about resilience in an increasingly volatile operating environment.

The study also identified gaps in organisational risk governance. Around 17% of respondents said they are unclear about what happens after new risks are identified, while 25% lack clarity on escalation triggers. Only about one in five organisations reported having a very clear understanding of how global and local response processes align. According to the report, these shortcomings can result in duplicated efforts, inconsistent responses, increased operational pressure and a higher risk of employee burnout during crises.

James Wood, Security Director for Northern Europe at International SOS, said the biggest obstacles are no longer recognising risks but making timely decisions once they arise. He noted that many organisations remain structured for stability rather than volatility, making it difficult to act quickly when rapid responses are essential.