THE government is exploring artificial intelligence (AI) as a tool to improve how agencies track public programs and assess whether these are delivering results, but its adoption will require safeguards on data quality, transparency and accountability.
Department of Economy, Planning, and Development Secretary Arsenio M. Balisacan said AI could help address delays, fragmented reporting and manual processing that can limit the government's ability to generate timely information for decision-makers.
The technology, however, should not be allowed to determine public-sector decisions on its own, Balisacan said during the 13th Monitoring and Evaluation (M&E) Network Forum in Mandaluyong City on Sept. 9.
“Our task, therefore, is not to surrender judgment to automation. It is to determine where AI can support our work, where it may mislead us, and what safeguards should govern its use,” he said.
The forum, held in collaboration with the United Nations Development Programme (UNDP), brought together government agencies, development partners, academic and research institutions, civil society and the private sector to examine how AI-supported evidence could be used in policy and implementation decisions.
Balisacan said AI could help make public-sector monitoring more efficient, but its contribution should first be tested in actual institutional settings.
He also warned that AI systems could reproduce errors and biases already present in the data and processes on which they are built.
“Technology alone does not produce credible evidence,” Balisacan said, adding that data become evidence through sound methods, careful interpretation and an honest account of their limitations.
Human review therefore remains necessary, he said, so officials can understand how an AI-supported system arrived at its results, examine the basis of its findings and remain accountable for decisions based on them.
The Philippine Development Plan 2023–2028 identifies research, technology, innovation, digitalization and good governance as part of the country's transformation agenda.
An example of how AI is already being applied to government functions was presented by the Philippine Space Agency (PhilSA), which showcased its Data Analytics Technologies and Operations Services for Space Data (DATOS) and Disaster, Hazard, and Risk Mapping (DHaRMa) initiative.
The initiative combines AI and satellite data for applications such as landslide, oil spill and fire monitoring, as well as mapping burned areas, ashfall and flood extent. Satellite imagery can also support rapid assessment of areas affected by flooding.
“AI can support monitoring and evaluation. But it must also become a subject of monitoring and evaluation,” Balisacan said.
He said agencies should assess whether AI-supported systems achieve their intended results and whether their benefits justify their costs and risks.
“When we leave this gathering, let us measure progress not by how much AI we use, but by how wisely and responsibly we use it,” Balisacan said.