OVER 92 percent of Philippine organizations use artificial intelligence. Fewer than one in eight have hired someone to govern it. The Swarm Philippine AI Report 2025, based on a nationwide survey of 175 organizations, has a term for it: “vibe-based” AI use. Fifty-seven percent cite talent scarcity as their top barrier to scaling. So where are the jobs?
At AXA Philippines, a job listing for a Compliance Officer for Data Privacy and Artificial Intelligence sits under the Office of the President. The role includes regulatory reporting to the National Privacy Commission, privacy impact assessments, AI compliance assessments, breach monitoring and stakeholder training. Applicants are expected to have three to seven years of experience in a regulated financial environment and familiarity with generative AI, machine learning and big data analytics. The application closes on June 25, 2026.
I looked at the listing closely because it shows what AI governance actually involves day to day: monitoring incidents, coordinating breach response and localizing a global company’s compliance framework to Philippine law. AXA placed the role in its compliance department, which says something about where organizations think AI risk lives — not in IT.
RealPage, a property technology company headquartered in Texas, has offices in Pasig City. Right now, it is hiring for three roles in Manila along the same lines: AI Compliance Specialist, AI Governance Analyst and an AI Ops Analyst whose job involves designing human-in-the-loop processes for the company’s AI agents. Abroad, the titles get more senior: Head of Responsible AI and AI Ethics Subject Matter Expert. We’re not there yet, but the direction is the same.
Policy is moving in the same direction. Department of Education Order 003, s. 2026, says AI should complement teaching, not replace teacher judgment. It cannot be the sole basis for grading or major academic decisions. Education Secretary Sonny Angara was candid about the reason: AI use in classrooms had moved faster than the rules. Schools are now expected to inventory AI tools, classify them by risk level and establish approval paths for higher-risk applications. Somebody will have to run that process, from compliance coordinators to AI registry officers, across division and regional offices nationwide.
At Ateneo de Manila University, policy requires that data privacy and intellectual property protections be upheld in all AI use. Researchers, not AI systems, bear ultimate responsibility for their work.
Then there is the Grok episode. On Jan. 16, the Department of Information and Communications Technology blocked Grok nationwide. The AI chatbot, developed by xAI, had been generating nonconsensual deepfakes of real people, including minors. The agency invoked the Cybercrime Prevention Act. Five days later, after xAI made content moderation changes, DICT lifted the ban. Regulators moved first and negotiated after.
DICT’s broader “AI for the People” initiative and the planned National AI Center for Research and Innovation under the Department of Science and Technology are shaping what government AI should look like going forward. The direction is toward systems with human oversight built in, not bolted on.
Governance roles like the ones at AXA and RealPage are already being filled in other countries. Bias and Fairness Analysts identify discrimination in machine learning models. AI Security Experts work with standards like ISO/IEC 42001 and monitor synthetic identity fraud. Globally, iProov’s Threat Intelligence Report 2025 recorded a 2,665-percent increase in native virtual-camera attacks and a 300-percent rise in face-swap attempts. For Philippine organizations handling identity verification, the risk is already here.
Jack Madrid, president of the IT and Business Process Association of the Philippines, projects the industry will hit 1.97 million full-time employees and $42 billion in revenue by year’s end. Routine customer queries are increasingly going to AI agents. What lands on the human agent’s desk now are the calls where someone is frustrated, confused or dealing with something the bot cannot read. Who decides when the AI hands off, and who answers for it when the handoff fails?
A KPMG analysis published on its Philippine site in March found that nearly 68 percent of organizations expect to scale AI by the end of 2026, but far fewer feel confident they have the governance structures to manage the risk. An International Labour Organization research brief published in February estimates 12.7 million Filipino workers are exposed to generative AI — the highest rate among Asean countries with comparable data — with women facing disproportionately higher exposure in administrative and financial roles.
The AXA listing and the RealPage postings are early signals. Only 12 percent of organizations have someone in that seat. The job boards suggest more are coming. The other 88 percent will need to catch up.
