SDAIA — Saudi Data and Artificial Intelligence Authority

The Saudi Data and Artificial Intelligence Authority was established in August 2019 by Royal Decree to lead Saudi Arabia’s national data and AI agenda. SDAIA occupies a unique position in the global AI governance landscape: it combines regulatory authority over data governance with operational responsibility for AI development, creating a single entity that sets the rules and builds the technology simultaneously.

SDAIA’s National Strategy for Data and AI, announced in October 2020 and known as ASPIRE, outlines a multi-phased approach to establishing Saudi Arabia as a data-driven economy. The strategy targets 20,000 trained AI specialists, 300 AI-driven startups, and over 20 billion dollars in data and AI investment. Priority sectors include education, healthcare, energy, mobility, and government — the domains where AI deployment can generate the largest socioeconomic impact.

The authority’s most significant technical contribution is the ALLaM language model, developed by SDAIA’s National Centre for Artificial Intelligence. The ALLaM project mobilized 16 government entities to contribute training data, assembled 400 subject matter experts for testing, and created the world’s largest purpose-built Arabic training corpus at 500 billion tokens. This institutional mobilization demonstrates the advantage of sovereign AI development: no private company could access the government data or command the expert resources that SDAIA deployed.

SDAIA also operates the national data governance framework, setting policies for data collection, storage, sharing, and protection that affect all organizations operating in Saudi Arabia. The Personal Data Protection Law, which SDAIA helped develop and now enforces, governs how personal data is processed in AI systems — including the Arabic language models that process citizen data for government services.

ASPIRE Strategy Framework

SDAIA’s National Strategy for Data and AI, known as ASPIRE, operates across six pillars with 66 specific targets. The strategy is phased for progressive ambition: national urgencies by 2025 (establishing foundational capabilities and governance), competitive advantage by 2030 (achieving sector-specific AI leadership), and global leadership post-2030 (positioning Saudi Arabia as a top-tier AI power).

The 20,000 AI specialists target addresses the human capital constraint that limits AI adoption across the Kingdom. Saudi Arabia’s education system, while producing large numbers of graduates, has historically underinvested in AI-specific curricula. SDAIA’s training programs, university partnerships, and scholarship initiatives aim to build the technical workforce that AI deployment at national scale requires.

The 300 AI-driven startups target reflects recognition that government-led AI development, while essential for sovereign capabilities, cannot alone generate the diverse applications needed for a data-driven economy. The ALLaM Challenge — offering SAR 1 million in prizes for innovative Arabic AI applications — represents one mechanism for catalyzing startup development. The GAIA Accelerator, a $1 billion regional AI accelerator launched through collaboration with New Native and NTDP, provides another.

The $20 billion+ investment target has been substantially exceeded. Saudi Arabia saw $9.1 billion in AI funding through 70 deals in 2025 alone. The Saudi Cabinet’s designation of 2026 as the Year of AI, alongside the count of 664 AI companies operating in the Kingdom, demonstrates that the investment target has catalyzed activity beyond initial projections.

Transition to HUMAIN

The May 2025 launch of HUMAIN as a PIF-backed national AI company redefined SDAIA’s operational scope. While SDAIA retains regulatory authority over data governance and AI strategy, the commercial execution of AI development — building models, deploying infrastructure, acquiring partnerships — transferred to HUMAIN. This organizational split separates policy-making from commercial execution, addressing a structural tension inherent in SDAIA’s original dual mandate.

HUMAIN inherited the ALLaM model family, the National Centre for Artificial Intelligence’s research capabilities, and the commercial relationships with IBM and Microsoft that SDAIA had established. HUMAIN’s mandate to become the world’s third-largest AI provider (behind the US and China) reflects an ambition scale that government authority structures are not designed to deliver.

The relationship between SDAIA and HUMAIN parallels the relationship between regulatory authorities and national champions in other sectors. SDAIA sets the rules — data governance, privacy protection, AI ethics frameworks — while HUMAIN plays by them at commercial speed. This separation enables HUMAIN to sign $23 billion+ in deals since launch, build 11 data centers, and establish partnerships with xAI, Adobe, NVIDIA, AMD, and AWS without the bureaucratic constraints that government operations entail.

Global AI Governance Position

SDAIA’s combination of regulatory authority and technical capability positions Saudi Arabia as a significant voice in global AI governance discussions. The authority’s ranking of first in public sector AI adoption globally demonstrates that SDAIA’s governance framework enables rather than constrains AI deployment. Saudi Arabia’s 14th-place ranking in the 2025 Global AI Index reflects the country’s rapid ascent from a position of limited AI capability to global significance within six years of SDAIA’s establishment.

SDAIA participates in international AI governance forums where its experience — governing AI development at national scale while supporting rapid commercial deployment — provides a model that other countries study. The tension between data protection (PDPL compliance) and AI capability development (maximizing training data availability) is a challenge every nation faces, and SDAIA’s approach to balancing these imperatives within an Islamic governance framework adds a unique perspective to global discussions.

The Personal Data Protection Law that SDAIA enforces affects every Arabic AI system operating in Saudi Arabia. AI models processing citizen data — chatbots handling government service inquiries, healthcare AI processing patient information, financial AI analyzing banking transactions — must comply with PDPL requirements for data residency, consent management, and processing limitation. HUMAIN’s decision to serve ALLaM from Saudi-based infrastructure directly addresses PDPL compliance, demonstrating how regulatory requirements shape deployment architecture.

Priority Sector Deployments

SDAIA’s five priority sectors — education, healthcare, energy, mobility, and government — define the domains where AI investment is directed. Education applications include Arabic-language tutoring systems, automated assessment, and curriculum personalization. Healthcare applications encompass diagnostic support, treatment recommendation, and administrative automation. Energy applications target optimization of Saudi Arabia’s oil and gas operations and the growing renewable energy sector. Mobility applications address traffic management, autonomous vehicle readiness, and public transportation optimization. Government applications span citizen services, regulatory compliance, and administrative efficiency.

Each sector presents specific Arabic AI requirements. Education demands Arabic NLP capable of analyzing student writing across formal and dialectal registers. Healthcare requires medical Arabic terminology understanding and patient privacy protection. Energy uses technical Arabic specific to petrochemical and renewable energy domains. Government requires the sovereign data access and institutional knowledge that ALLaM’s training data provides.

SDAIA’s Organizational Evolution and HUMAIN Transition

SDAIA’s organizational trajectory reflects the maturation of Saudi Arabia’s AI ambitions from regulatory authority to commercial execution. Initially established as the national data and AI authority — a regulatory and strategic body — SDAIA evolved to include the National Centre for AI (NCAI), which developed ALLaM. The subsequent creation of HUMAIN as a separate government-owned company under PIF management represents the institutional separation of regulation (SDAIA) from commercial execution (HUMAIN).

This separation follows a governance pattern used across Saudi economic sectors: regulatory authorities set rules while national champions execute commercially. SDAIA retains authority over data governance (including PDPL implementation), AI ethics frameworks, workforce development strategy, and international AI policy engagement. HUMAIN assumes responsibility for commercial AI products (ALLaM), infrastructure (data centers), and ecosystem development (venture fund, accelerator programs).

The dual-entity structure enables each organization to optimize for its mandate. SDAIA can develop regulations without commercial conflicts of interest. HUMAIN can pursue commercial growth without regulatory constraints. The coordination between entities ensures that regulation supports rather than hinders AI development — a balance that SDAIA’s dual role previously required managing within a single organization.

SDAIA’s Data Collection and Curation Infrastructure

SDAIA’s authority over national data governance provided unique access to training data for ALLaM’s development. The mobilization of 16 public entities to contribute data — under SDAIA’s data governance framework — represents a data collection capability that no private organization could replicate. Government documents spanning decades of Saudi administrative history, regulatory frameworks, educational materials, healthcare records (anonymized), and infrastructure documentation were assembled into the 500-billion token Arabic training dataset.

The 400 subject matter experts engaged by SDAIA to test ALLaM through over one million prompts represent additional institutional capability. These experts — spanning medicine, law, engineering, education, Islamic studies, and Arabic linguistics — provided domain-specific evaluation that validated model accuracy across the professional domains where incorrect AI outputs carry consequences. This expert-in-the-loop evaluation methodology has been adopted by other Arabic AI development programs, confirming its value for culturally and professionally aligned model development.

The 300 Arabic books selected for inclusion in ALLaM’s training data were curated for linguistic quality and domain coverage — a selection process that required Arabic literary and academic expertise beyond what automated quality filtering provides. This manual curation, combined with automated processing of larger data volumes, created a training corpus that balances quality (expert-curated content) with quantity (web-scale data collection).

SDAIA’s International Positioning and Global AI Governance

Saudi Arabia’s 1st place ranking in public sector AI adoption globally — achieved under SDAIA’s strategy execution — positions the Kingdom as a model for government AI deployment that other nations study. The ranking reflects extensive AI integration across Saudi government services, from citizen interaction platforms to internal administrative automation. The 14th place ranking in the 2025 Global AI Index indicates rapid ascent from limited AI capability to global significance within six years of SDAIA’s establishment.

SDAIA participates in international AI governance forums, including bilateral AI cooperation agreements and multilateral discussions on AI safety, ethics, and development standards. Saudi Arabia’s experience — deploying AI at national scale within an Islamic governance framework that addresses cultural, religious, and social considerations absent from Western AI governance models — provides a perspective that enriches global AI governance discourse. The AraTrust benchmark’s evaluation of trustworthiness across eight culturally relevant dimensions exemplifies the evaluation frameworks that SDAIA’s governance approach has influenced.

SDAIA’s Workforce Development and Training Programs

SDAIA’s workforce development programs address the talent pipeline that Arabic AI ecosystem growth requires. Training academies, university partnerships, and professional certification programs build AI capability across Saudi Arabia’s existing workforce. The Data and AI Academy provides specialized training for government employees deploying AI in public services. University partnerships with Saudi institutions integrate AI curriculum into engineering, computer science, and data science programs.

The NSDAI/ASPIRE target of 20,000 AI specialists by 2030 requires both domestic training and international recruitment. SDAIA coordinates with Saudi universities on AI program expansion, with international institutions on research collaboration, and with technology companies on applied training programs. The Year of AI 2026 designation amplifies these workforce programs by elevating AI to national priority status — creating cultural momentum toward AI careers that complements formal training infrastructure.

SDAIA’s certification programs establish professional standards for AI practitioners in Saudi Arabia. Data governance certifications ensure that professionals handling training data and model deployment understand PDPL requirements and data protection best practices. AI ethics certifications address the cultural and regulatory considerations specific to Arabic AI deployment — including content sensitivity, cultural alignment, and Islamic governance principles that Western AI ethics frameworks do not address.

The $9.1 billion in 2025 AI funding and 664 AI companies in Saudi Arabia create employment demand that absorbs the growing AI workforce. SDAIA’s workforce programs produce graduates who enter positions across the AI ecosystem — from foundation model development at HUMAIN to applied AI deployment at startups to government AI integration across Saudi ministries. This workforce distribution network ensures that SDAIA’s training investment generates returns across the entire Saudi AI economy rather than concentrating talent at a single organization.

SDAIA’s Legacy and Institutional Significance

SDAIA’s significance extends beyond its direct outputs to its institutional role as the architect of Saudi Arabia’s AI ecosystem. Every component of the Kingdom’s AI infrastructure — HUMAIN, ALLaM, the data center program, the venture fund, the accelerator, the workforce programs — exists within the strategic framework that SDAIA created. The institution’s combination of regulatory authority, data governance responsibility, and strategic planning capability provides the coordinated approach that Arabic AI development requires — an approach that fragmented institutional structures in other countries struggle to achieve.

Related Coverage

• MENA AI Companies — Full company directory
• Arabic LLMs — Foundation model coverage
• ALLaM — National Model — Model analysis
• HUMAIN Profile — Commercial successor
• SDAIA Strategy — Detailed strategy analysis
• Project Transcendence — $100B initiative
• Year of AI 2026 — National designation
• MENA Startup Ecosystem — Investment landscape