(Credit: Sami Nakib via AI)
Artificial Intelligence (AI) in Lebanon is often misunderstood. Some see it as a magical fix for all of the country's woes, a swift shortcut to bypass the deep reforms or hard work required to build strong institutions. Others see it as an imminent threat to jobs or reduce it to handy tools like text or image generators.
The framing of AI as either a threat or a magic wand masks the profound shifts this new technology is bringing about. As thoroughly argued in a new white paper by Dr. Fatima K. Abu Salem from the American University of Beirut (AUB) and Dr. Wissam Saade from Saint Joseph University, no technology, no matter how advanced, can ever replace a strong society built on solid institutions, shared values and collective effort.
Published as a working paper at AUB’s Issam Fares Institute for Public Policy and International Affairs, the report is more than just an academic document — it’s a manifesto with urgent recommendations to strengthen Lebanon’s five-year AI strategy, LEAP, launched by Technology and AI Minister Kamal Shehadeh, who also serves as Minister of the Displaced.
LEAP — short for Launch, Apply, Advance, and Promote — is a national roadmap to make Lebanon a regional hub for artificial intelligence within five years. “We want to integrate the latest technologies to create government services powered by AI,” Shehadeh told L’Orient-Le Jour in April, describing a vision that focuses on improving the efficiency and accuracy of public services. The strategy’s first phase aims to establish the legal and institutional foundation, aiming to rank Lebanon among the top 50 AI users worldwide by 2030.
"For LEAP to move beyond lofty aspirations, it must confront Lebanon’s reality as a ‘data desert’ facing a ‘talent exodus,’ and embrace an AI approach co-created with the broader public," Abu Salem told L’Orient Today.
Abu Salem and Saade lay the groundwork for a future Sociotechnical Observatory, which would serve as a public space to examine how technology and society interact and to explore new ways of thinking — informed by the local context — about AI in Lebanon’s political, social and economic life.
Abu Salem and Saade’s white paper highlights a key insight: AI isn't just a technical tool; it's a deeply "socio-technical" project. This means that for AI to succeed in Lebanon, strategies for its adoption must allow for "co-evolution," meaning that the technology develops alongside Lebanon's unique local conditions, values and systems. This approach and the resulting recommendations will be discussed at the "AI Retreat in Aley," part of the "AI for Lebanon" conference (July 21-26, 2025).
AI models developed elsewhere aren’t neutral, and importing them without adaptation won’t magically make them fit. Any national strategy must integrate AI into Lebanon's broader social fabric, including its labor market, infrastructure and power dynamics. "Without this deep integration, LEAP risks becoming just another initiative focused on technological adoption rather than sustainable transformation," says Saade.
At the heart of this socio-technical approach is Responsible AI (RAI), which involves designing and using AI in ways that are ethical, fair, transparent and aligned with human values. It is supported by Trustworthy Machine Learning (TML), a framework that guarantees AI systems are understandable, secure and capable of adapting to real-world conditions.
"TML also emphasizes the importance of having a clear understanding of an AI’s level of confidence in its predictions — essential in critical fields like healthcare — and promotes Machine Learning for Small Data, a key method for data-scarce contexts like Lebanon," says Abu Salem.
The white paper proposes three main pillars to guide this socio-technical transformation:
“Collective intelligence” reimagines AI not as a surveillance tool but as a way to broaden the scope of what distinct groups can achieve when their data is pooled together, often through crowdsourcing. In Lebanon, this means citizens can act as "human sensors," providing timely, low-cost data that fills institutional gaps and encourages community ownership and inclusion.
“Data democratization” refers to making data accessible and easy to use for everyone in an organization, not just IT or data experts. It's about breaking down barriers so that more people can engage with data and fostering a culture where decisions are based on facts, not just intuition.
For LEAP, this would mean moving beyond fragmented, ad-hoc data practices to a scalable and integrated approach that empowers the general public to engage with data.
“Decolonizing AI” challenges Lebanon’s dependence on imported AI models, which are shaped by Western values and power structures. The white paper warns against "data colonialism" and "algorithmic coloniality," where local data is exploited and inequality is reinforced. For Abu Salem and Saade, this power dynamic can be addressed by reclaiming control over how data is collected and categorized, as well as by whom it is collected.
They also highlight the high energy requirements of large AI systems, which is a serious concern in a country facing severe electricity shortages. That’s why the authors advocate for Micro-Model Alternatives: smaller, low-energy AI systems that can operate on basic devices, making AI more sustainable and better aligned with Lebanon’s realities.
"We want to explore how Lebanon can both create value from AI, through better decision-making, demand prediction and smarter services, and capture it by translating innovation into economic and strategic gains," says Saade.
The paper warns that, if misused, AI could harm development, worsen inequality and waste resources. In some cases, traditional, non-digital solutions may be more effective, challenging the tech-first approach that prioritizes flashy projects over genuine impact.
It also calls for rethinking human labor: Since AI handles routine tasks, the remaining work will require critical thinking, ethics and adaptability. This shift necessitates updates in labor policy, education and protections for hybrid work arrangements.
For LEAP to succeed, Abu Salem and Saade emphasize that it must center on the human experience and be grounded in Lebanon’s social reality, providing clear guidance for an ethical and inclusive AI strategy.
Aoun hails agreement between Washington and Tehran after call from Iran FM