With AI, we should rethink our economic development model

This is a matter where my research and my personal circumstances, coming from a family of call centre and virtual assistant workers, intersect, so forgive me for the first-person tonal break:

In 2023, I woke up to OpenAI's announcement of GPT-4 and early demonstrations of tool-using capabilities in a customer service context. I immediately penned this op-ed for Rappler on the implications of widespread agentic AI adoption for the economic structure of the Philippines.

My capstone research project at Yale-NUS College used a combination of ethnographic methods, qualitative GIS, and industry data analysis to detail how work-from-home (WFH) in the business process outsourcing (BPO) sector exposed the dependence of the Philippine urban development model on one industry. Some may try to explain the resulting shifts in economic geography as a temporary pandemic-induced phenomenon. In my conclusion, I reflect on how a longer-lasting shift in BPO office space demand may shift urbanisation and infrastructure development patterns in the Philippines permanently, with less growth in established city centres like Manila and Cebu. (And this is before we even consider the current acceleration of data centre and telecommunication infrastructure construction in the country, as well as resulting ecological impacts.)

I will admit: I'm struggling to come up with a simple, fair takeaway for this post. It's easy to collapse the narrative into the Philippines being "in danger" with rising AI adoption in businesses. But revenue and headcount data continue to look strong. And while customer experience chatbots are becoming more commonplace, there is still a long way to go before generative AI will be trusted with the full gamut of tasks performed by BPO workers in the Philippines, who are working for clients across industries such as healthcare, accounting, and law.

I'm left wondering, however: why are Filipino industry analysts often too comfortable with using these caveats to justify complacency? Why is the conclusion often "we have no reason to worry yet," rather than "what do we do when it's time to worry?"

What would it take for our country to reconsider the legacy of the past few decades of structural transformation- and to critically consider the paths forward?

(One thing is for sure: this is an exciting, if also exhausting, time to be a researcher studying the relationship between technology and urbanisation.)

Read the original op-ed below.