AI for All Begins with a State 5.0 Structure

In the public sphere, artificial intelligence is a technology that impresses and sometimes frightens. Yet it is becoming something else entirely: the very material with which the State will build its services, the new brick and mortar of public action, resting on the black gold of the 21st century, data.

This material is quietly settling into ordinary interactions as simple as a benefits application, a medical appointment, a tax return or a request for a daycare space. But AI, like any masonry, will only hold on solid foundations: coherent data infrastructures, governed from one end of the value chain to the other. The question is simple: “Will citizens be able to understand where their data goes, have an error corrected, and know who remains accountable for an AI-assisted decision?”

The sustainable digital value chain

Canada is right to want an ambitious strategy, since it has the assets: research, talent, innovative companies and democratic institutions. But a government strategy cannot be limited to funding tools or accelerating productivity; that would be to confuse digital transformation with sustainable digital transformation. The former optimizes a portfolio of projects; the latter governs a value chain made up of sovereign infrastructures, data, capabilities, services, public value and resilience. In my doctoral work, I define this whole as the “sustainable digital value chain”¹, in which each of the six links is indispensable. Weakness in a single one creates a technological, organizational or informational debt that ultimately falls back on the citizen². In other words, trust in an “AI for all” is decided less at the interface than in its tangible foundations.

Toward a State 5.0

This trust cannot rest on legal provisions alone: it requires an alignment between a legal foundation and a data architecture designed for pooling across public bodies³. This is the meaning of a “State 5.0”: an administration structured around data, human-centred, capable and sustainable, built for an uncertain and changing environment. Administrations that succeed with AI are distinguished not primarily by the technology they acquire, but by their organizational capacity to create value⁴.

It is therefore important to understand that a system plugged into a department’s applications creates value only if citizens derive a concrete benefit from it, such as better service, shorter delays, controlled public costs and protected data. If the State no longer knows who does what, where data flows and how a decision can be challenged, it cedes part of its control to external actors.

A person whose file is refused following an automated analysis should never be told merely that “the system recommended it, and that is that.” They should know the determining criteria, be able to request a human review, have an error corrected and obtain an intelligible explanation. AI can certainly speed up administration, but not at the expense of proximity or transparency.

Five essential questions for citizens

Canada’s strategy should set a simple requirement: no major public use of AI should be funded or deployed without a demonstration of measurable and auditable organizational capacity. That demonstration comes down to five questions:

Who is responsible? The department, the vendor, the integrator, the host, the operator and the subcontractors must be distinguished. Without this mapping, accountability disappears at the very moment citizens need it.

What happens to the data? Inputs, outputs, logs, metadata and developments carried out for a public mandate must be defined contractually. Public data, a strategic asset of the State, should never be used to train other models without explicit authorization.

How is the system verified? Transparency is not limited to announcing that an AI tool is in use. It requires traces, socio-technical audits, testing and correction mechanisms in order to optimize its costs per business task.

Is the organization ready? Before any acquisition, a public body should demonstrate its capacity to oversee the tool: data governance, cybersecurity, in-house skills, contract management and accessible recourse. It is internal capabilities, more than the purchase itself, that determine success.

Can control be regained? The administration must be able to suspend a system, change vendors, recover its data, revert to a human process and maintain essential services, all while controlling its costs. This resilience presupposes sovereign infrastructures comparable in importance to the great projects of the 20th century.

Sustainable innovation and agentic AI

These requirements are not hostile to innovation; they make it sustainable and socially acceptable. They foster a healthier market, built on security, interoperability and accountability, and democratize good governance for SMEs and municipalities. Advanced states such as Estonia, where interoperability is enshrined in law, or Switzerland with its shared open models, show that success depends less on a central, technology-centred body than on distributed capabilities.

The urgency grows with agentic AI: systems capable not only of analyzing but of acting in autonomous operational workflows. It is no longer enough to assert that a human remains “in the loop”; we must specify which human, at what moment, with what power to interrupt and with what duty to account.

First and foremost an institutional issue

An ambitious Canadian strategy should aim for more than rapid adoption of “AI for all”: it should offer citizens a shared public AI resting on a solid legal foundation, cross-cutting governance and sustainable organizational capabilities. The administration of tomorrow can be more efficient while remaining traceable, accountable and contestable, and can call on private partners without losing its institutional memory, its data or its digital sovereignty. For beyond the technologies, the issue is first and foremost institutional: the states that will get the most out of AI will not be those that have acquired the most algorithms, but those that have invested in their internal capabilities, their data governance, their sovereign infrastructures and their democratic control over automated decisions, in keeping with the principles of sustainable digital maturity. In other words, a national strategy should measure organizational maturity as much as technological performance.

Finally, “AI for all” will truly begin the day every citizen can say: I understand enough to trust, my data remains protected, and I have recourse when a machine-assisted human gets it wrong. If AI is to become the new brick and mortar of the State, Canada’s ambition should not only be to remain a research leader, but to build a State 5.0 whose foundations we control, in order to develop a sovereign AI on a sustainable value chain.

Statement on the use of AI tools

Generative AI tools were used as an aid in structuring the text to the competition’s guidelines. The author personally wrote the original content and revised, verified and approved the final content, which is based on her original research.

Notes

¹ Vallée, S. (2026). Chaîne de valeur du numérique durable [Sustainable digital value chain]. 2nd Annual Symposium of the Technology, Sustainability and Society Research Chair, Université du Québec à Chicoutimi, June 23, 2026.

² Teece, D. J. (2007). Explicating dynamic capabilities: The nature and microfoundations of (sustainable) enterprise performance. Strategic Management Journal, 28(13), 1319–1350; Kattel, R., & Takala, V. (2021). Industrial and Corporate Change, 30(4). [Reference to be confirmed.]

³ Vallée, S., & Ebata, R. (2026). IA et cadre normatif québécois : indice de rupture documentaire et exigences de la transformation numérique durable [AI and Quebec’s normative framework: Documentary disruption index and requirements of sustainable digital transformation]. Revue Organisation et Territoires (peer-reviewed). Forthcoming, fall 2026.

⁴ Mergel, I., Edelmann, N., & Haug, N. (2019). Defining digital transformation: Results from expert interviews. Government Information Quarterly, 36(4), 101385.

References

Bharadwaj, A., El Sawy, O. A., Pavlou, P. A., & Venkatraman, N. (2013). Digital business strategy: Toward a next generation of insights. MIS Quarterly, 37(2), 471–482. https://doi.org/10.25300/MISQ/2013/37:2.3

Elkington, J. (1997). Cannibals with forks: The triple bottom line of 21st-century business. Capstone.

Innovation, Science and Economic Development Canada. (2026). Canada’s National Artificial Intelligence Strategy: AI for All. Government of Canada.

Kattel, R., & Takala, V. (2021). Industrial and Corporate Change, 30(4). [Reference to be confirmed.]

Mergel, I., Edelmann, N., & Haug, N. (2019). Defining digital transformation: Results from expert interviews. Government Information Quarterly, 36(4), 101385. https://doi.org/10.1016/j.giq.2019.06.002

Moore, M. H. (1995). Creating public value: Strategic management in government. Harvard University Press.

Porter, M. E. (1985). Competitive advantage: Creating and sustaining superior performance. Free Press.

Treasury Board of Canada Secretariat. (2019, updated). Directive on Automated Decision-Making. Government of Canada.

Teece, D. J. (2007). Explicating dynamic capabilities: The nature and microfoundations of (sustainable) enterprise performance. Strategic Management Journal, 28(13), 1319–1350. https://doi.org/10.1002/smj.640

Vallée, S., & Ebata, R. (2026). IA et cadre normatif québécois : indice de rupture documentaire et exigences de la transformation numérique durable [AI and Quebec’s normative framework: Documentary disruption index and requirements of sustainable digital transformation]. Revue Organisation et Territoires (peer-reviewed). Forthcoming, fall 2026.

Vallée, S. (2026). Chaîne de valeur du numérique durable [Sustainable digital value chain]. 2nd Annual Symposium of the Technology, Sustainability and Society Research Chair, Université du Québec à Chicoutimi, June 23, 2026.

Verhoef, P. C., Broekhuizen, T., Bart, Y., Bhattacharya, A., Qi Dong, J., Fabian, N., & Haenlein, M. (2021). Digital transformation: A multidisciplinary reflection and research agenda. Journal of Business Research, 122, 889–901. https://doi.org/10.1016/j.jbusres.2019.09.022

International Organization for Standardization. (2023). ISO/IEC 42001:2023 Information technology — Artificial intelligence — Management system. ISO.