Capiz State University

A successful dialogue must move beyond abstract ethics to enforceable, practical rules that empower frontline conservationists with deployable tech—turning AI from a theoretical tool into a lifeline for ocean health. Without this specificity, the dialogue risks becoming a hollow policy exercise. It must deliver tangible, actionable outcomes focused on practical implementation and ecosystem protection. Establish an AI Governance Framework Agreement on mandatory standards recognized by all nations, Localized Sovereignty & Capacity Building, and clear/shared KPIs linking AI governance to conservation outcomes.

• Safe, secure and trustworthy AI
• AI capacity-building
• Social, economic, ethical, cultural, linguistic and technical implications of AI
• Open-source software, open data and open AI models

Safe, Secure, and Trustworthy AI - Prioritized to prevent cyberattacks. AI Capacity Building - Essential to empower nations and communities (coastal communities) with local technical ownership. Without this, systems risk becoming dependency-driven on external vendors, undermining sovereignty and sustainability. Capacity building turns theoretical frameworks into deployable, community-led solutions for long-term monitoring. Social, Economic, Ethical Implications - Prioritized to avoid exclusionary or exploitative practices. Open-Source Software, Data, and Models - Critical for collaborative adaptation. Let small nations customize solutions without vendor lock-in, while open data ensures transparency in marine conservation metrics. This fosters global knowledge-sharing, enabling low-resource countries to build resilient, context-specific monitoring systems.

Data sovereignty, Transboundary Accountability Fragmentation, neglecting systems-level fragility in climate-adapted, energy-limited, and geopolitically complex nations.

Undermining enforcement of AI governance. Data sovereignty gaps allow foreign entities to extract ecological data without consent, while capacity constraints prevent local teams from maintaining systems during connectivity outages. Climate-driven ecosystem shifts further destabilize static AI models, exacerbating operational failures. Opportunities, however, are transformative. Open-source toolkits empower communities to customize systems to address threats, reducing reliance on expensive vendors. Transboundary cooperation—via frameworks like the Global Dialogue—harmonizes standards across nations, turning fragmented systems into coordinated enforcement networks. Prioritizing AI capacity building ensures local community-led deployments, embedding ownership and adaptability.

The Global AI Dialogue can catalyze meaningful international cooperation on AI governance by transforming abstract principles into actionable frameworks in shared ecological spaces. Its most significant role lies in bridging fragmented national policies through: Standardizing cross-border Edge AI protocols: By harmonizing data sovereignty rules (e.g., local processing mandates) and security requirements. The Dialogue could jointly establish enforceable standards. Enabling knowledge-sharing ecosystems: Prioritizing open-source toolkits would allow coastal nations to co-develop and deploy systems without vendor lock-in, fostering collaborative innovation rather than siloed efforts. Creating enforcement synergies: Harmonizing AI governance with existing treaties (e.g., UNCLOS) would integrate into collective enforcement mechanisms, ensuring tools are used consistently—turning technology into a shared conservation asset.

The Global AI Dialogue should strategically integrate with and elevate three key initiatives while adding unique value: Builds on its ecosystem-monitoring frameworks but lacks AI governance. The Dialogue could formalize real-time AI protocols, turning them into enforceable action. Added value: embedding AI governance into existing frameworks ensures tools directly support conservation goals. The Dialogue should mandate transboundary integration of AI. The Dialogue can establish localized adaptation mandates, requiring open-source models to include climate-resilient features, ensuring open tools evolve with ecosystem needs instead of becoming static, vendor-dependent assets.

The Global AI Dialogue should leverage stakeholder-driven, practical collaboration to ensure its relevance for AI Governance. Key contributions and structural recommendations: Stakeholder Roles: Governments: Prioritize binding commitments for national AI standards (e.g., mandatory local processing) and fund capacity-building for all nations, especially coastal nations. NGOs/Conservationists: Share field-tested use cases and advocate for Indigenous knowledge integration into AI models. Tech Firms: Develop open-source, low-power AI tools and commit to data sovereignty protocols. Indigenous Communities: Lead co-design of AI tools to respect ecological traditions, integrating various algorithms. Format & Structure Recommendations: Multi-Session Workshops: Start with plenary consensus-building on core principles, followed by technical working groups to draft actionable frameworks. Field Validation Sprints: Host in-situ testing of AI tools to ensure operational relevance. Tiered Follow-Up: Establish national action plans for signatories, with quarterly progress reviews led by a dedicated "AI Tech Task Force" to track implementation.

Currently underrepresented voices in global AI governance include Indigenous coastal communities, small island developing states (SIDS), Local Government Units from remote areas, and fisher folks. These groups face systemic exclusion despite holding deep ecological knowledge and direct experiences. Indigenous communities, stewards of ancestral domains, are often denied control over AI-generated data, while SIDS lack the capacity to adapt high-cost AI tools. To include them effectively: Co-design mandates: Require Indigenous knowledge integration into AI training and fund community-led data sovereignty initiatives. Resource equity: Prioritize SIDS' access to low-bandwidth AI toolkits through Global Dialogue's funding streams—ensuring no nation is locked into proprietary systems. Operational representation: Embed local experts and Indigenous leaders in technical working groups to define field-test protocols (e.g., validating AI alerts against traditional monitoring practices).

The AI Dialogue must move beyond traditional forums to create hyper-local, dynamic engagement where stakeholders co-define solutions for AI Governance. Conduct knowledge circles with indigenous elders, and the like.