Researchers, policymakers, and technologists will gather in Quezon City this week as AI Safety Diliman hosts its AI Governance Accelerator Cohort 1 Demo Day, highlighting locally grounded research projects tackling the growing societal and political implications of artificial intelligence in the Philippines.

Set for May 24 at SevenSpace along Maginhawa Street, the event will feature interdisciplinary projects examining how AI systems intersect with public discourse, labor, law, and governance. Organizers say the initiative reflects a broader push to develop AI governance frameworks that are sensitive to Philippine realities rather than merely adapting models from Western countries.

One of the featured studies explores how large language models respond to politically charged Philippine issues, using the impeachment of Sara Duterte as a case study. The project examines whether AI systems take explicit political positions, avoid them altogether, or produce inconsistent responses depending on the framing and language of prompts. Researchers said the study raises broader concerns surrounding information reliability, political neutrality, and the governance risks tied to generative AI systems operating in multilingual environments such as the Philippines.

Another project focuses on the country’s business process outsourcing sector, one of the Philippines’ largest industries and among the most exposed to automation pressures. The “BPO AI Worker Dialogue Lab” investigates how AI tools are reshaping workflows, productivity targets, and workplace monitoring systems. Through workshops and worker-centered dialogue, the initiative seeks to develop governance approaches that reflect the day-to-day realities faced by Filipino workers adapting to AI-integrated workplaces.

Legal and institutional responses to AI will also take center stage during the event. A separate research initiative titled “From Principles to Black Letter Law: Embedding AI Safety in Legal and Institutional Frameworks” examines how abstract AI safety principles can be translated into enforceable governance mechanisms. Using the Philippines as a primary case study, the project explores how Southeast Asian countries that largely consume rather than develop frontier AI technologies can still build meaningful regulatory safeguards against systemic and cross-border AI risks.

According to organizers, the Demo Day underscores the growing recognition that AI governance debates must move beyond purely technical discussions and account for local institutional conditions, labor structures, political environments, and educational systems.

The event is supported by Kairos, BERI, and AI Safety Network. Interested participants may register through the official event page at Luma Event Page.