1. Sovereignty is being built through infrastructure, not restriction

The Philippines is approaching AI sovereignty through a combination of domestic infrastructure investment and differentiated controls over data based on sensitivity.

Executive Order 119 divides government data into Restricted Access and Open Access categories. More sensitive information remains within national jurisdiction, while other data can be hosted on secure commercial cloud infrastructure. At the same time, the Department of Science and Technology (DOST) is expanding its AI data centre and GPU capacity 26-fold by 2028, with the stated objective of strengthening the country’s ability to develop and run AI applications on locally controlled infrastructure.

This makes sovereignty partly a capacity question. The Philippines is investing in the infrastructure needed to retain domestic capability for sensitive workloads, while allowing appropriate use of commercial cloud infrastructure for other data and applications.

The resulting model could give the country greater flexibility in how AI infrastructure is developed and deployed. At the same time, the definition of “restricted” data becomes particularly important. How those categories are interpreted and applied will ultimately determine how much of the AI environment needs to remain under domestic control.

2. AI & semiconductors are being positioned as export opportunities

A fast-tracked government incentive package targets USD 110 billion in AI and semiconductor exports by 2030. That is an ambitious target and signals a broader attempt to position AI as an economic growth opportunity rather than primarily a tool for improving government or enterprise productivity.

The direction also represents a shift in how the Philippines could build its position in the technology value chain. The country has historically built its international economic position around business process outsourcing and services rather than hardware or deep-tech production. Linking an AI export target with semiconductor development suggests an effort to move into higher-value activities and capture more of the technology value chain.

Whether the USD 110 billion target can be reached by 2030 is less important than what the target signals about government priorities: AI is being treated as part of the country’s future export and industrial strategy.

The capabilities that develop around this ambition will be critical, including data centres, semiconductor facilities, engineering talent, AI infrastructure, and supporting services. The pace and coordination of these investments will help determine whether the export ambition can translate into a broader technology base.

3. Fragmented AI initiatives are moving towards a common strategy

The Private Sector Advisory Council has proposed a National AI Implementation Taskforce to bring together what it describes as fragmented AI efforts across government, industry, and academia. Its recommendations span workforce development, infrastructure, and policy, alongside a proposal to release approximately USD 11.2 million from the Higher Education Development Fund for AI upskilling programs in 2026.

The significance is the recognition that fragmentation itself has become a constraint. Multiple agencies and institutions can pursue useful AI initiatives, but without coordination, investment, skills development, infrastructure planning, and policy can move in different directions.

The fact that this push is coming from the private sector is also notable. PSAC is calling for greater coordination and a more coherent national approach rather than relying on individual government agencies to develop their own programs.

That also highlights a current gap in the Philippines’ AI governance structure: there is not yet one institution with clear end-to-end ownership of the country’s AI agenda. The eventual role and authority of the proposed Taskforce will therefore matter. If it has a clear cross-agency mandate, budget, and ability to coordinate implementation, it could help consolidate existing initiatives. If it becomes primarily another advisory layer, it could add to the institutional complexity it is intended to address.

4. Government AI is being directed towards disaster resilience

Some of the most visible government AI applications in the Philippines are being directed towards disaster preparedness and resilience.

The country has allocated USD 16.4 million to Project NOAH for AI-powered flood forecasting and real-time hazard assessment. DepEd’s LIGTAS+ platform is also being used to monitor storm, heat, and volcanic risks across more than 47,000 public schools. Alongside these initiatives, DICT and Google Cloud are deploying Gemini Enterprise to help citizens navigate government services, including disaster-related information, in local languages.

The emphasis on disaster resilience reflects the Philippines’ operating environment. Monsoons, typhoons, flooding, and volcanic activity create challenges that are specific to the country’s geography and population. AI is being applied to problems where better prediction, earlier warnings, and faster access to information can have a direct public impact.

This also has implications for how government AI gains public acceptance. Disaster-response applications have a different political and social proposition from automation designed primarily to reduce administrative costs or headcount. Their benefits can be measured through improved preparedness, faster response, and potentially fewer lives and assets at risk.

That makes disaster resilience a potentially important entry point for public sector AI in the Philippines. If these applications demonstrate clear benefits, they could also help build institutional and public confidence for broader uses of AI across government services.

5. AI access is extending beyond large organisations & major cities

The Philippines is also putting attention into who gets access to AI. Around one million public school teachers are being given access to Microsoft Copilot for lesson planning, assessment, and administrative work. DICT is running AI skills and connectivity programs for students and MSMEs in regions such as Ilocos, combining public Wi-Fi with basic AI training and partnerships involving AWS, Google, and Microsoft.

A new national Cybersecurity Council is also intended to strengthen the cyber foundations supporting the country’s wider digital development, while the government is exploring digital governance cooperation with Poland.

Individually, these initiatives are relatively modest compared with major infrastructure or investment announcements. However, they address an important part of the AI equation: access and capability beyond the largest organisations and the capital.

Infrastructure investment and AI export ambitions will have limited impact if the workforce does not have the skills to use the technology. Extending access to teachers, students, and MSMEs creates a broader base from which AI adoption can develop, particularly in regions outside Metro Manila.

There is also a practical reason to focus on these groups early. AI capability can otherwise become concentrated among large enterprises, technology companies, and organisations with the resources to experiment. Regional skills and connectivity programs provide a way to broaden participation, although access alone will not guarantee meaningful adoption. The harder question is whether small businesses and other users have the time, skills, and business incentives to turn that access into sustained productivity or new services.