Strategic Argument and Areas of Debate
Middle powers face a structural paradox where pursuing total technological autarky ensures economic isolation, yet unmanaged interdependence risks the permanent erosion of sovereign strategic agency within computationally mediated governance. The viability of AI sovereignty for states like Türkiye therefore depends on reconciling domestic indigenisation with the global reality of highly concentrated semiconductor and compute supply chains.
Executive Summary
The paper argues that Türkiye is uniquely positioned to achieve AI sovereignty through a model of managed interdependence that leverages its established defence-industrial indigenisation expertise. By moving beyond the National AI Strategy 2021–2025 toward the Integration phase defined in the 2026 Presidential Annual Programme, the Republic of Türkiye aims to embed machine learning into state capacity while navigating US-China rivalry and EU AI Act regulatory standards. Successful execution requires the Ministry of Industry and Technology and the Digital Transformation Office to address structural gaps in semiconductor access and the 1 GW data centre capacity target by 2030.
Analytical Framework and Key Drivers
Six-Layer Sovereign AI Stack: This framework disaggregates national capacity into distinct layers from Energy and Physical Infrastructure to Applications and Services, preventing the strategic fragility caused by over-investing in visible layers while neglecting foundational dependencies.
Managed Interdependence Strategic Model: Rather than pursuing impossible full-stack self-sufficiency, states categorise systems into three sensitivity tiers to balance domestic control over critical national infrastructure with the pragmatic use of global hyperscale platforms for commercial applications.
Regulatory Capacity as Sovereignty: Building domestic expertise and engaging with the OECD AI Principles and ITU standards allows middle powers to exert normative influence, ensuring global governance frameworks do not become exclusively US-centric or Chinese-centric.
Defence-to-Civilian Technology Transfer: The indigenisation of military capabilities, such as the Steel Dome and Kızılelma programmes, provides a sovereign foundation of proprietary datasets and engineering talent that can be systematically re-tasked for civilian industrial modernisation.
Strategic Assessment & Empirical Findings
- McKinsey estimates that sovereign AI requirements will influence 30 to 40 per cent of total global AI spending by 2030, representing a market valued between 500 and 600 billion USD.
- Türkiye currently maintains a data centre capacity of 250 MW, which is less than 0.5 per cent of national electricity consumption, necessitating a projected 10 billion USD investment to reach its 1 GW target by 2030.
- The European Union has committed approximately 200 billion EUR to the AI Continent Action Plan, yet it remains structurally dependent on US hyperscalers for its underlying compute infrastructure.
- The global AI compute market is dominated by only nine public cloud providers from the United States and China, who collectively control roughly 70 per cent of the industry.
- Türkiye’s workforce development goals aim to produce 50,000 AI professionals with graduate-level specialisation by 2030 to mitigate brain drain and capture the 4 per cent global GDP uplift projected by the IMF.
- The 2026 Presidential Annual Programme marks a shift toward Integration, treating AI as a cross-cutting instrument of state capacity rather than a siloed innovation tool.
Geopolitical Trajectories & Policy Risks
- US export controls on advanced accelerators like the Nvidia H100 create a permanent structural risk for middle powers, potentially forcing a choice between technological stagnation or alignment with the United States-China axis.
- The Republic of Türkiye faces a critical implementation gap if the draft AI Law (TBMM No. 2/2234) remains unpassed, as regulatory uncertainty deters the institutional and foreign investment needed for the 2030 Industry and Technology Strategy.
- Failure to align capital instruments with specific stack layers—such as providing patient, long-term capital for energy infrastructure versus venture pathways for applications—will lead to systematic resource misallocation within the Turkish AI ecosystem.
Critical Policy Questions & Responses
Question 1 Why does the concept of managed interdependence challenge traditional notions of technological autarky for middle powers?
Answer: Traditional autarky is rendered unviable by the extreme scale economies and supply chain concentrations found in the semiconductor and foundation model layers. Managed interdependence allows countries like Türkiye to preserve strategic agency by securing domestic control over sensitive tier-one defence and intelligence data while pragmatically leveraging global platforms for non-critical commercial functions.
Question 2 How does Türkiye’s defence-industrial base provide a unique advantage for its broader sovereign AI ambitions?
Answer: Unlike purely commercial strategies, the Turkish defence sector has generated decades of proprietary operational datasets and domestic engineering talent through programmes like the Steel Dome. This institutional knowledge creates a sovereign data asset that can be transitioned into civilian applications like infrastructure inspection and disaster search-and-rescue via the TSSK framework.
Question 3 What are the consequences of Türkiye’s reliance on European Union regulatory alignment for its domestic AI governance?
Answer: By aligning the draft AI Law with the EU AI Act, Türkiye ensures high levels of interoperability with its primary trading partner while avoiding the status of a rule-taker. However, this approach requires the Republic of Türkiye to maintain national discretion through the National AI Registry to ensure that compliance burdens do not stifle domestic innovation or national technology goals.
Question 4 What risks emerge if the gap between Türkiye’s data centre capacity targets and energy infrastructure planning remains unaddressed?
Answer: While the 2030 Industry and Technology Strategy targets a 1 GW capacity, current energy forecasts from the Ministry of Energy and Natural Resources may undercount the tripling of demand. Without public anchor investment and green energy co-location, the 10 billion USD in projected private investment may fail to materialise, leaving Turkish startups dependent on external cloud providers.
Key Actors and Systemic Dynamics
- Republic of Türkiye → Depends on → Global semiconductor supply chains
- Ministry of Industry and Technology → Coordinates with → Digital Transformation Office (CBDDO)
- TÜBİTAK BİLGEM → Enables → Indigenous semiconductor design
- ASELSAN → Supports → Secure microelectronics access
- EU AI Act → Shapes → Turkish draft AI Law (TBMM No. 2/2234)
- United States → Constrains → Advanced AI chip exports
- NATO → Influences → Turkish defence-industrial standards
- 2026 Presidential Annual Programme → Accelerates → AI-enabled state capacity
- Public Data Space (Kamu Veri Alanı) → Enables → Sovereign Turkish LLM development
- Cognitive Diplomacy framework → Strengthens → Turkish normative influence
