Poke Bowl

From an SME owner's perspective, the first Global Dialogue on AI Governance will be successful if it delivers practical outcomes for companies that want to grow responsibly across borders. My company aims to expand internationally through franchising and we are highly interested in adopting AI tools in our operations. For businesses like ours, the key issue is predictability. If we invest in AI systems today, we need confidence that these tools can also be used by franchise partners in other countries without major legal barriers, conflicting rules or costly redesigns. A successful dialogue would therefore support more interoperable, risk-based governance and reduce regulatory fragmentation between markets. Another critical issue is data security and responsible use. In practice, employees and individuals often test different AI tools even when company policy does not allow confidential business data to be entered into them. SMEs need clearer answers on accountability: how can a business owner be sure that data entered into AI systems will not leak, be reused or later become retrievable through other prompts? This concern is even greater with aggregator platforms that offer access to multiple LLMs in one place. These services are easy for anyone to launch and easy for users to access, but it is often unclear who is responsible for data handling, what safeguards exist and how trustworthiness is verified. For this Dialogue to matter to businesses, it should not focus only on frontier developers or governments. It should also address the real operational concerns of SMEs: cross-border usability, vendor accountability, data protection, employee behavior and ethical deployment in everyday business settings. I believe SME voices are essential in this discussion, because we face the implementation risks directly and can contribute concrete, practical experience that can help make AI governance workable in the real economy.

Safe, secure and trustworthy AI;Transparency, accountability, and human oversight;Social, economic, ethical, cultural, linguistic and technical implications of AI;Interoperability of governance approaches;

I selected these four priorities because they best reflect the real concerns of an SME that wants to adopt AI responsibly and expand across borders through franchising. Interoperability of governance approaches is essential because we need confidence that AI systems adopted in one country can also be used by franchise partners in other markets without conflicting legal requirements or costly redesign. Safe, secure and trustworthy AI is a priority because businesses need stronger assurance that company data entered into AI systems will not be leaked, reused or exposed through other prompts or third-party platforms. Transparency, accountability and human oversight matters because SMEs need clarity on who is responsible for data handling, what safeguards vendors must provide and how businesses can maintain control over employee use of AI tools in practice. Social, economic, ethical, cultural, linguistic and technical implications of AI is also relevant because SMEs experience AI not as an abstract policy issue, but as something that directly affects day-to-day operations, staff behavior, customer trust and ethical business conduct across different markets. Together, these four priorities capture the practical, cross-border and trust-related issues that matter most to smaller businesses using AI in the real economy.

Yes. A few important cross-cutting issues are not fully visible in the listed themes, especially from an SME perspective. One is cross-border enforceability in practice. Interoperability is listed, but SMEs also need practical mechanisms that make compliance portable across countries, so businesses do not have to reinterpret rules market by market. Another is vendor and platform-chain accountability. Many companies now use not just one AI provider, but a chain of providers, plugins, API layers, resellers and multi-model platforms. Who is responsible for data protection, auditability and failure when several actors are involved. A third is SME-usable governance. Many frameworks are written in ways that large firms can manage, but smaller businesses need simple, affordable, implementable standards, templates and procurement guidance. Without this, governance may exist formally but remain inaccessible in practice. Also important is employee-level AI use and shadow AI. Even with internal policies, staff may test public or third-party tools using company information. This creates a governance gap between formal rules and day-to-day business reality. Another emerging issue is trust in AI aggregators and multi-model access platforms. These services are growing quickly, but questions remain about data routing, storage, subcontracting, jurisdiction and liability. This area deserves more explicit attention.

AI governance gaps create uncertainty around cross-border use, data security, vendor accountability and practical SME compliance. Estonia has a strong digital policy tradition and continues to position AI adoption as an economic opportunity, while the EU AI Act is becoming the main legal framework shaping how AI is used and governed. For SMEs, that creates both momentum and uncertainty: adoption is encouraged, but implementation details, responsibilities across the value chain and simplified compliance for smaller firms are still evolving. As a business operating in a highly digital environment and aiming to expand internationally through franchising, we need confidence that AI systems adopted in one market can also be used in others without conflicting requirements or costly redesign. We also see a growing gap between internal company policies and real employee use of public or third-party AI tools, especially where sensitive business data may be entered into systems with unclear retention, reuse or accountability arrangements. For SMEs in our sector, the main challenge is not only adopting AI, but doing so in a way that is secure, transparent, operationally realistic and workable across jurisdictions.

The AI Dialogue can advance international cooperation by helping countries and stakeholders move toward more interoperable, practical and predictable approaches to AI governance. For SMEs, its value lies in reducing fragmentation between markets, clarifying accountability across the AI value chain and supporting common expectations on safety, transparency and responsible cross-border use. It can also ensure that international AI governance reflects real business conditions by including the operational experience of SMEs, not only the perspectives of governments or large technology companies.

From an SME perspective, the Dialogue could bring real value by elevating issues that are still under-addressed in many existing forums: cross-border usability of AI systems, vendor-chain accountability, data protection in everyday business use and governance of multi-model AI platforms. It can also help ensure that international AI governance is shaped not only by governments and large technology firms, but also by smaller businesses and real-economy users who face implementation challenges directly.

From an SME perspective, one practical recommendation is to reserve dedicated space for real-economy sectors such as hospitality, retail, health, education, and manufacturing. The added value of the Dialogue would then be not only broad inclusion, but the ability to connect global governance discussions to operational realities that smaller businesses face every day.

Generally, I think global AI governance would benefit from hearing not only from those who design or regulate AI, but also from those who must adopt it, manage its risks and live with its consequences in practice.

In my view, meaningful engagement is more likely when the Dialogue goes beyond formal statements and allows participants to bring practical experience into the discussion. One useful format could be small, moderated thematic roundtables with balanced participation from governments, business, civil society and technical experts. These are often more dynamic than large plenaries and can make it easier for less prominent voices to contribute. It may also be valuable to include case-based discussions built around real examples. For instance, participants could react to short scenarios on cross-border AI use, data protection, vendor accountability or employee use of public AI tools. This would help move the conversation from abstract principles to practical governance questions. It could be useful to end sessions with brief outcome-oriented summaries that capture points of convergence, open questions and practical next steps. In that way, engagement would not only be dynamic, but also more likely to produce usable input for future cooperation.

From a practical business perspective the most effective approaches are those that make AI use more transparent, accountable, and workable for smaller companies. Useful examples include clear employee rules on AI use, stronger vendor transparency on data handling, standard contractual safeguards, staff training and simple risk-based guidance that SMEs can realistically apply. Effective governance should not only set principles, but also help smaller businesses adopt AI responsibly, reduce uncertainty and use AI across borders with greater confidence.