Enterprise vendors are tightening a 'triple bind'—lock-in, bundling and scale—to extract higher prices and reduce CIOs' negotiating room, a squeeze set to deepen as AI centralizes data and models; CIOs must adopt modular architectures, multi-vendor sourcing and stronger contract terms to preserve sovereignty.

How to Achieve Digital Sovereignty

van Giffen, Benjamin, Brenner, Barbara, Brenner, Walter · August 30, 2026 · Journal of the Association for Information Systems

openalex commentary low evidence 7/10 relevance Summary only summary available; pdf_status=paywall Source PDF

Structured author observations

Linked only from stored provider relations; the raw author line above is never matched by name.

OpenAlex

Latest observation: September 3, 2026

van Giffen, Benjamin provider ID
Brenner, Barbara provider ID
Brenner, Walter provider ID

Vendors use a ‘triple bind’ of lock-in, bundling and scaling to weaken enterprise digital sovereignty, raising prices and exit costs—a dynamic that AI-driven data and model consolidation is likely to intensify unless CIOs pursue defensive strategies.

Citation observations

Cumulative provider counts captured on specific dates; providers are never combined.

0 cumulative citations

OpenAlex · Observed September 3, 2026

View corpus context

CIOs consistently report that software vendors are dramatically increasing license costs while reducing negotiation flexibility. Increased prices result from the erosion of digital sovereignty, limiting an organization’s ability to control its digital infrastructures. Our research reveals how vendors leverage the “triple bind” (lock-in, bundling and scaling) to create insurmountable dependencies, which will likely intensify further in the AI era. This leaves enterprises with little room to negotiate or exit. Though there is no silver bullet, we recommend four strategies that CIOs can adopt to regain control.

Summary

Main Finding

Vendors are using a “triple bind” of lock-in, bundling and scaling to erode digital sovereignty, enabling sustained and growing price increases and sharply reduced negotiation flexibility for enterprise buyers. This dynamic is likely to strengthen in the AI era, leaving CIOs with little room to negotiate or exit unless they adopt deliberate defensive strategies.

Key Points

Triple bind explained:
- Lock-in: technical, contractual and data dependencies make switching costly or infeasible.
- Bundling: vendors combine products, services and platforms to capture more spend and obscure marginal prices.
- Scaling: economies of scale and network effects amplify vendor market power as AI platforms consolidate data, models and customers.
Consequence: rising license and consumption costs, fewer effective concessions in procurement, and higher exit costs for enterprises.
AI intensifies the problem by increasing value of integrated datasets/models, accelerating platform consolidation, and creating new proprietary layers (models, fine-tuning, inference services) that deepen dependence.
No single remedy exists; CIOs must pursue multiple, complementary approaches to recover bargaining power and protect digital sovereignty.

Data & Methods

Evidence base: synthesis of CIO reports and complaints, market pricing observations, documented vendor contracting practices, and case examples of procurement outcomes.
Analytical approach: conceptual framing of vendor tactics as the “triple bind,” supported by qualitative examples and trend analysis showing vendor consolidation and price trajectories in software and cloud services.
Limitations: research relies primarily on reported experiences and market-level observation rather than a single randomized, causal study; specifics of each enterprise’s exposure depend on architecture, sector regulation and vendor mix.

Implications for AI Economics

Market structure: AI-driven network effects and data concentration increase barriers to entry and strengthen incumbents’ pricing power, shifting surplus toward vendors.
Firm behavior and investment: higher vendor prices and exit costs alter CIOs’ investment calculus—favoring lock-in-friendly vendor solutions unless countermeasures are taken—potentially reducing competition-driven innovation at the enterprise level.
Welfare and regulation: growing vendor power raises concerns about allocative efficiency, pricing transparency, and the need for policy interventions (data portability, interoperability standards, procurement rules).
Distributional effects: smaller firms and public sector organizations with limited leverage will be disproportionately affected, widening capability and cost gaps.

Practical strategies (recommended) 1. Design for modularity and portability: adopt modular architectures, APIs and open standards to reduce switching costs and isolate vendor-controlled layers. 2. Diversify sourcing and adopt multi-vendor/multi-cloud strategies: fragment dependence so no single provider can extract outsized rents; include open-source and alternative vendors in procurement pipelines. 3. Strengthen contract terms and procurement practices: insist on data portability, clear pricing formulas, caps/benchmarks for price increases, audit and SLA enforcement rights, and defined exit/transition support. 4. Build internal capabilities and coalition strategies: invest in in-house platform skills, leverage open-source AI where feasible, and pursue collective bargaining or industry consortia to increase negotiating leverage and influence standards/regulation.

Short-term priorities for CIOs: map vendor dependencies and data flows, quantify exit costs, renegotiate renewals with portability and pricing safeguards, and pilot modular replacements for the most strategic vendor lock-ins.

Assessment

Paper Typecommentary Evidence Strengthlow — The argument is based on synthesis of CIO reports, market observations and case examples rather than systematic causal analysis or representative, quantitative data; claims are plausible but not empirically validated for causal effect sizes. Methods Rigorlow — Methods rely on qualitative synthesis and illustrative cases without a clear sampling frame, systematic data collection, or counterfactual analysis; potential selection and reporting biases are not addressed. SampleQualitative synthesis drawing on CIO reports and complaints, market pricing observations, documented vendor contracts and procurement case examples; no single representative dataset or randomized/quasi-experimental sample specified and time/geography coverage not reported. Themesgovernance adoption GeneralizabilityFindings are enterprise-focused and may not apply to consumer or small-business markets, Reliance on self-reported CIO experiences risks selection and survivorship bias, Market structure and vendor behavior vary across industries, countries and regulatory regimes, Extent and impact of AI-specific consolidation will differ by firm architecture, data sensitivity and vendor mix, Observational, case-based evidence limits ability to generalize effect sizes or causal mechanisms

Claims (11)

Claim	Direction	Outcome	Confidence & Evidence	Details
Enterprise software and cloud vendors use a combination of lock-in, bundling, and scaling to increase their bargaining power over enterprise buyers. Market Structure	negative	Enterprise buyer bargaining flexibility	Reading fidelity high Study strength medium	not reported 0.06
Vendor lock-in raises the cost or feasibility barriers associated with switching providers. Market Structure	negative	Provider switching costs	Reading fidelity high Study strength medium	not reported 0.06
Bundling allows vendors to capture a larger share of customer spending and makes marginal prices less transparent. Market Structure	negative	Pricing transparency and vendor share of enterprise spending	Reading fidelity high Study strength medium	not reported 0.06
Economies of scale and network effects increase vendor market power as AI platforms consolidate data, models, and customers. Market Structure	negative	Barriers to entry and incumbent pricing power	Reading fidelity high Study strength medium	not reported 0.06
The vendor triple bind is associated with rising license and consumption costs, fewer effective procurement concessions, and higher enterprise exit costs. Firm Revenue	negative	Enterprise software and cloud purchasing costs and exit costs	Reading fidelity high Study strength medium	not reported 0.06
AI intensifies vendor dependence by increasing the value of integrated datasets and models, accelerating platform consolidation, and adding proprietary layers such as models, fine-tuning, and inference services. Automation Exposure	negative	Enterprise dependence on AI vendors	Reading fidelity high Study strength medium	not reported 0.06
Higher vendor prices and exit costs can shift enterprise investment toward lock-in-friendly solutions and potentially reduce competition-driven innovation. Innovation Output	negative	Enterprise investment choices and competition-driven innovation	Reading fidelity high Study strength low	not reported 0.03
Smaller firms and public-sector organizations with limited bargaining leverage are likely to experience disproportionately greater capability and cost gaps. Inequality	negative	Organizational technology costs and capabilities	Reading fidelity high Study strength low	not reported 0.03
Modular architectures, APIs, and open standards can reduce switching costs and isolate vendor-controlled layers. Organizational Efficiency	positive	Switching costs and vendor dependence	Reading fidelity high Study strength low	not reported 0.03
Multi-vendor and multi-cloud sourcing can fragment dependence and reduce the ability of any single provider to extract outsized rents. Market Structure	positive	Vendor bargaining power and procurement costs	Reading fidelity high Study strength low	not reported 0.03
Contract provisions covering data portability, transparent pricing formulas, price-increase caps or benchmarks, audit rights, service-level enforcement, and exit support can help protect enterprise bargaining power. Governance And Regulation	positive	Procurement bargaining power and exit feasibility	Reading fidelity high Study strength low	not reported 0.03