The Technopolitical Order Debate: Rival Visions of AI and Power, and Türkiye's Position

Source: Millî İstihbarat Akademisi (National Intelligence Academy, Ankara), "Dünya Siyasetinde Teknopolitik Düzen Tartışmaları ve Türkiye" ["Technopolitical Order Debates in World Politics and Türkiye"], Teknopolitik Çalışmalar Dizisi-1, July 2026.

I. EXECUTIVE ASSESSMENT

The source report, issued by Türkiye's National Intelligence Academy under the auspices of its president, Prof. Dr. Talha Köse, argues that the global debate over artificial intelligence has moved decisively beyond questions of technical capability. At stake, in the report's framing, is who will hold authority once AI becomes load-bearing infrastructure for state power, economic hierarchy, and social legitimacy — and how a mid-sized, geographically exposed state such as Türkiye should position itself within that transition.

The report's core argument rests on four linked claims: frontier AI has become a determinant of state capacity rather than merely an efficiency tool; the international debate has crystallized around three schools of thought — a security-and-power-maximizing "technorealism," a safety-and-governance-centered "technopragmatism," and a dignity-and-legitimacy-centered "technohumanism"; none of the three offers a complete strategy for a state in Türkiye's position; and the appropriate response is therefore synthesis — building strategic capacity, safe institutional design, and democratic legitimacy simultaneously, while explicitly contesting the hierarchical civilizational logic embedded in the dominant realist current.

The report is notable less for its technical content than as evidence that a state intelligence institution is treating the AI-governance debate as directly relevant to defense and intelligence planning, and is explicitly positioning Türkiye's own civilizational and normative tradition as a competing pole in that debate rather than a passive subject of it.

Several of the report's underlying assumptions merit scrutiny. It assumes continued concentration of power among a small number of frontier AI developers, producing a capability hierarchy the report compares loosely to the post-1945 nuclear order — an assumption drawn largely from the technorealist camp the report otherwise critiques. It treats Türkiye's centralized administrative tradition as double-edged: a coordination asset in crisis response, but a multiplier of the centralizing tendencies already inherent in AI systems. And it assigns real strategic value to normative and civilizational argument — the report's least evidenced claim, resting on inference about future diplomatic traction rather than demonstrated effect, and best read as an aspirational element of strategy rather than a settled forecast.

II. STRATEGIC CONTEXT

The report situates its argument within four converging structural shifts, which it treats as interacting rather than independent.

Power transition and AI sovereignty

States and firms that control frontier model development, compute at scale, and proprietary data are gaining disproportionate influence over defense targeting, public administration, and information environments. The report defines sovereignty in this context as the capacity to audit, modify, or replace foreign-controlled AI embedded in critical state functions without unacceptable disruption cost.

Infrastructure, minerals, and the compute stack

The report presents a three-tier stack — advanced chips, energy, data infrastructure and robotics at the base; technical capacity in the middle; political and economic outcomes at the top — in which control at any lower tier constrains what is achievable above it.

Data governance as the binding constraint

Across many middle-income and regionally exposed states, the report argues, technical capability is now outrunning institutional capacity to govern it. This capability–governance gap, rather than raw compute deficit, is presented as the more diagnostic indicator of near-term AI-related vulnerability, including for Türkiye itself.

III. THREE COMPETING MODELS

The report's central analytical contribution is a typology of three schools competing to define the emerging technopolitical order.

Model A — Technorealism

This current seeks to reconstitute a tight coupling between state power and frontier technology, treating AI as a hard-power instrument comparable to nuclear weapons. The report anchors it in the circle around Palantir co-founder and CEO Alex Karp, Peter Thiel, Joe Lonsdale, Palmer Luckey, and Marc Andreessen, tracing its intellectual roots to René Girard and Leo Strauss and, at its most radical edge, to the neo-reactionary "Dark Enlightenment" current associated with Curtis Yarvin and Nick Land. The report's central criticism is that technorealism's implicit distinction between "functional" and "dysfunctional" cultures recodes older civilizational-hierarchy logics in technological language, and that its proponents — several of whom, the report notes, hold direct commercial stakes in defense AI contracts — blur the line between strategic analysis and commercial advocacy.

Model B — Technopragmatism

This current holds that powerful AI systems generate both large benefits and large risks simultaneously, and that safety must be constitutive of design rather than added after deployment. The report anchors it in Anthropic (citing Dario Amodei and Chris Olah), and situates it alongside the Center for AI Safety and Google DeepMind's Frontier Safety Framework. Its central limitation, per the report, is that it remains company-centered: the same firms defining "responsible" capability thresholds are simultaneously racing to build the underlying capability, leaving the question of democratic oversight largely unresolved.

Model C — Technohumanism

This current subordinates both state and market deployment of AI to human dignity, the common good, and social justice as binding constraints rather than one input among several. The report anchors it in Pope Leo XIV's 2026 encyclical "Magnifica Humanitas," UNESCO's AI ethics recommendation, and the work of Shoshana Zuboff and Ruha Benjamin. Its central limitation is dependence on voluntary coalition-building; the report notes it offers the strongest normative grounding of the three but the weakest institutional mechanism for translating that grounding into enforceable policy.

IV. COMPARATIVE ASSESSMENT

The report concludes that none of the three models is viable as a complete governing framework on its own. Technorealism is judged the most institutionally embedded in the near term, given its existing defense-industrial relationships. Technopragmatism is judged the most exportable in vocabulary terms, with its risk-tiering and red-teaming language already migrating into EU and NIST regulatory architecture. Technohumanism is judged potentially the most durable current over the long run if it succeeds in broadening its coalition and embedding itself in law and procedure rather than remaining exhortatory. Technorealism is additionally flagged as the most internally contradictory of the three: its stated defense of Western institutions sits uneasily against a governance mechanism that concentrates unaccountable authority in a narrow security-industrial elite. For middle powers generally, the report's conclusion is that no single model fits alone — the appropriate response is deliberate synthesis of elements from all three, not selection among them.

V. TÜRKİYE ASSESSMENT

Applying this framework to Türkiye, the report presents a genuinely mixed profile.

Strengths

The report cites a defense-industrial base with combat-tested experience integrating AI into autonomous systems, decision support, and targeting-relevant data pipelines; demonstrated public–private coordination through TÜBİTAK and the Ministry of Industry and Technology, including the TRUBA computing infrastructure and sector-specific funding calls; a large annual output of engineering and data-science graduates; and an expanding, if still fragmented, set of sectoral AI applications across the Ministries of Justice, National Education, Agriculture and Forestry, and Health.

Weaknesses

The report identifies a pronounced gap between the volume of data the Turkish state now holds and the maturity of the institutional architecture governing it — fragmented inter-agency data structures, inconsistent standardization, and limited transparency and explainability in AI-assisted decision-making, all of which the report argues can undermine institutional trust even where systems function technically.

Türkiye's near-term vulnerability, in the report's assessment, is not capability. It is the widening gap between how much data the state now holds and how little of it sits inside a governance architecture that can be meaningfully audited.

Dependence and governance risk

The report explicitly links compute and foundation-model dependency on foreign providers to Türkiye's earlier, costly delay in adopting nuclear technology, offering that experience as a cautionary precedent. It further argues that AI's inherent centralizing tendency compounds Türkiye's already centralized administrative tradition, raising the risk that local needs go unrecognized as oversight of concentrated power becomes harder to sustain.

VI. ANALYTIC JUDGMENTS

Reformulating the report's argument in estimative terms, with confidence levels reflecting the strength of the underlying evidence rather than certainty of outcome:

• Moderate confidence: no single model is likely to achieve unchallenged global dominance through the medium term, though technorealist logic is likely to exert disproportionate influence over material capability allocation given its existing institutional embeddedness.
• Moderate confidence: middle powers, including Türkiye, will increasingly articulate hybrid strategies borrowing from all three currents rather than aligning fully with any single bloc.
• Moderate confidence: data-governance maturity, rather than raw compute access, will likely prove the more binding near-term constraint for states in Türkiye's institutional bracket.
• Moderate-to-high confidence: the report's civilizational-hierarchy critique of technorealism is unlikely, on its own, to alter the material trajectory of frontier AI development absent a broader international coalition willing to operationalize it.

VII. ALTERNATIVE PERSPECTIVES

A technorealist reading would likely argue the report understates the urgency of great-power AI competition and overweights normative constraints relative to the pace of the security race it describes. A European regulatory reading would note that binding, risk-based regulation is already the EU's operational answer to the same coordination problem the report treats as unresolved. Read from a Global South perspective, the report's civilizational-hierarchy critique of technorealism would likely be found sympathetic but comparatively underdeveloped relative to its treatment of the other two models, and its own assumption of near-inevitable AI-driven power concentration — a premise it shares with the technorealist current it critiques — is itself open to challenge as a contestable trajectory rather than a foregone conclusion.

VIII. TÜRKİYE'S PROPOSED POLICY TRAJECTORY, PER THE REPORT

The report proposes a phased roadmap. In the short term (0–12 months): a mandatory inventory of AI systems in public-sector use, definition of use-case principles, construction of a risk-classification scheme, identification of high-risk systems, and preparation of AI-specific public procurement conditions. In the medium term (1–3 years): mandatory impact assessment, testing and certification infrastructure, data standardization, institutional separation of rule-making, implementation, and oversight functions, technical expert cadres within government, and controlled transfer of defense-sector AI capability into civilian domains such as disaster management, health logistics, and agriculture. In the longer term (3–5 years): a national public-sector AI governance framework, an articulated Turkish "AI doctrine," international technopolitical diplomacy, support for pluralistic international governance mechanisms, and continued strengthening of domestic compute and data infrastructure, including Turkish-language foundation models.

IX. FINAL ASSESSMENT

Bottom line: the report frames the global contest over AI's role in reordering state power, economic hierarchy, and human dignity as unlikely to converge on a single governing paradigm in the foreseeable future. It expects a hybrid landscape to persist, with technorealist logic dominating material capability allocation, technopragmatist frameworks supplying the dominant technical vocabulary for emerging regulation, and technohumanist critique retaining normative but not yet binding force. For Türkiye specifically, the report's central recommendation is that strategic capacity-building, safe institutional design, and democratic-legitimacy building must proceed in parallel rather than sequentially; it explicitly warns that pursuing capability without corresponding governance risks producing systems that function technically while eroding institutional trust.

Confidence in the report's longer-horizon judgments, particularly regarding whether binding or voluntary international governance mechanisms will ultimately prevail, remains moderate-to-low, given the still-unsettled state of international AI regulatory architecture as of mid-2026.