Generative AI in the Philippines’ BPO sector transforms tacit worker knowledge into proprietary assets, turning skilled cognitive labor into commodified inputs. The result is a new modality of accumulation — ‘cognitive dispossession’ — that channels value to platform owners and entrenches global inequalities.

Main Finding

Generative AI deployment in Philippine BPOs is enabling systematic capture and codification of workers’ tacit knowledge, reorganizing cognitive work around algorithmic rules, commodifying AI-augmented labour, and enclosing collective intelligence within proprietary systems. These processes operate together as a new modality of capital accumulation — “cognitive dispossession” — that redistributes value and power upward in global value chains and reinforces existing international hierarchies and inequalities.

Key Points

• Context: The Philippines hosts ~40% of the global customer experience workforce, making its BPO sector a strategic case for studying GenAI impacts on globally integrated knowledge work.
• Tacit knowledge capture: Workers’ know-how and contextual judgment are being extracted, formalized, and encoded into AI/automation systems.
• Codification and algorithmic governance: Cognitive tasks are reframed as rule-following or model-driven processes; decision-making becomes constrained by algorithmic protocols.
• Commodification of augmented labour: Worker contributions are transformed into data, models, and standardized outputs that can be packaged and sold as products or services.
• Enclosure of collective intelligence: Shared team practices, coordination routines, and local knowledge are internalized into proprietary platforms and IP, reducing workers’ claim over the value they collectively generated.
• Distributional consequences: Value and bargaining power shift toward firms and platform owners (often outside the local economy), exacerbating inequality across the global value chain.
• Conceptual contribution: Frames these interlinked dynamics as cognitive dispossession — a world-scale process of value extraction from embodied, situated human knowledge enabled by AI.

Data & Methods

• Empirical base: Qualitative study drawing on interviews with knowledge workers employed in Philippine-based BPO firms.
• Approach: Interpretive, grounded analysis of workers’ accounts to identify mechanisms of knowledge capture, work reorganization, commodification, and enclosure.
• Scope/limitations: Rich, context-sensitive insights into processes and power shifts; not a quantitative causal estimate. Findings are transferable to other globally integrated service sectors but should be tested with complementary quantitative and comparative methods.

Implications for AI Economics

• Value capture and rents: AI systems enable firms to internalize and commodify formerly dispersed human knowledge, concentrating economic rents in the hands of platform/firm owners. Measuring AI-driven rents requires tracking encoded knowledge and ownership of derived IP/data assets.
• Labour markets and bargaining power: Codification reduces skill rents tied to tacit expertise and can weaken workers’ leverage, potentially depressing wages or altering career ladders in service-exporting economies. Collective bargaining and new institutional arrangements may be required to protect workers’ claims.
• Global inequality and GVCs: AI-mediated cognitive dispossession can reinforce core–periphery asymmetries by transferring value upstream (to model owners and lead firms), altering where and how value is created within global value chains.
• Productivity measurement and welfare: Standard productivity metrics may misattribute gains from AI to firms rather than factoring in dispossessed human contributions, complicating welfare assessments and policy responses.
• Policy and governance needs: Consider interventions on data/knowledge ownership, IP rules, platform transparency, taxation of AI-derived rents, worker data rights, and support for upskilling/transition policies. Antitrust and labor regulations may be needed to address enclosure of collective intelligence.
• Research agenda: Quantify the scale of tacit-knowledge capture and its effects on wages and employment; map ownership of AI-encoded knowledge across GVCs; evaluate policies (data rights, co-ownership, taxation) to rebalance value distribution.