Leading the AI Revolution: 2026 Trends and Insights from Saudi Arabia & UAE

Artificial intelligence (AI) may be advancing rapidly, but the ability to explain how it works remains a challenge, particularly with Generative AI (GenAI) and large language models (LLMs). One organization tackling AI interpretability is Anthropic: an AI research, product, and safety company with whom Deloitte collaborated to develop this report.

Anthropic’s approach to interpretable AI promises to yield a better understanding of how AI systems work internally. However, the inability of human users to understand how AI is making the decisions or generates the content. AI still poses significant hurdles to scaling AI safely, making interpretability even more crucial for operational performance, risk management, and compliance with AI regulations—particularly in regulated industries.

Aspects of interpretability

As we explore the essential role of AI interpretability, several limitations present themselves, particularly technical challenges, operational obstacles, regulatory framework, and the evolving role of autonomous AI agents.

The future of interpretable AI

As AI systems continue to evolve toward autonomous decision-making, with minimal human oversight, AI interpretability will become not only a matter of compliance but a fundamental requirement for deploying increasingly complex and independent AI systems.

Organizations that proactively address this challenge by prioritizing interpretable models and transparent processes will be better positioned to leverage the transformative potential of AI. Deloitte’s collaboration with Anthropic not only helps organizations unlock this potential but also underscores our commitment to maintaining trust and accountability through responsible AI.

Middle Eastern regulators and boards are already treating AI interpretability as a hard requirement, not a “nice addition to have,” especially in financial services, public sector, and critical infrastructure where transparent GenAI decisions cannot be reconciled with emerging governance expectations. Countries such as the UAE, Qatar and Saudi Arabia are hard-coding transparency, traceability, and human oversight into their national AI charters and adoption frameworks. This raises the bar for explainability in any large-model deployment touching citizens or national assets. 

Dr. Aleksei Minin, Director at Deloitte AI Institute Middle East said that clients are increasingly asking not only how good the model operates, or if it can perform the required task more efficiently compared to a human, but if we can also prove why it works, to auditors, regulators, Sharia and ethics committees, as well as the public. As a result, interpretability tooling, documentation, and model-risk governance are becoming first-class design criteria in the region’s GenAI programs, on par with accuracy, latency, and cost. We do see some evolving tools that utilize GenAI more as an interface between humans and better-defined machine learning systems. An example of that is digital twins of different assets or processes, that allow the building of hybrid systems, being more reliable and transparent compared to pure GenAI systems. Nevertheless, the work Anthropic is doing is extremely interesting and important to foster GenAI adoption.