Towards a Socio-technical and Responsible AI Transformation in Lebanon : Guardrails from Socio-materiality and Trustworthy Machine Learning for Data Deserts

Abstract

As Lebanon embarks on a long overdue data and AI transformation, the daunting challenges and setbacks it has continuously grappled with raise substantially founded concerns around its ability to navigate the associated opportunities and risks, in an increasingly volatile and ambiguous world. In this working paper, we propose forward an ambitious manifesto and a suite of recommendations for complementing the five-year strategy (LEAP) launched by the ministry of state for Technology and AI in Lebanon. Whether among the elites or the general public, the question of artificial intelligence remains, in a country like Lebanon, shrouded in a jungle of myths and confusion. AI is perceived at times as an absolute threat to employment, and at others as a magical solution to all the country's problems. Often, it is reduced to limited applications such as text or image generation. This narrow view limits our ability to grasp the scope and depth of ongoing AI transformations happening worldwide. In Lebanon, as in other countries facing structural collapse, AI is frequently seen as a miraculous shortcut: a means to avoid institutional labor, deep reform, and long-term educational investment. Yet, no technological tool, however advanced, can replace a coherent and dynamic social, political, and economic ecosystem. Such an ecosystem relies on institutions, shared values, and sustained collective effort: elements that technology alone cannot create or sustain. In light of all that, this working paper aims to augment a plethora of channels by which to support the LEAP roadmap using a socio-technical and responsible AI parlance, against a backdrop of talent exodus. How can a data desert such as Lebanon progress robustly and reliably to become an AI hub? What evolving governance and ethical risks need to be navigated with the advent of GenAI and Agentic AI? How do we ensure we are creating value from AI whilst preventing value destruction in other non-AI systems? And how should we revise labor policies, understand human labor in light of required human agency, and safeguard human workers, in the AI age? For all intents and purposes, this manifesto prioritizes the need to render such ambitions credible and attainable, so we can move beyond AI proofs-of-concept towards AI deployment across our public and private sectors, robustly and responsibly.