High-impact economies (net negative): India, Philippines, Poland, Romania — countries heavily dependent on BPO, outsourced services, and offshore development. India alone could see 2-3 million BPO jobs at risk by 2028.
Moderate-impact economies (mixed): US, UK, Germany, Japan — advanced economies where AI creates and destroys jobs simultaneously, with net impact dependent on retraining policies.
Low-impact economies (limited exposure): Sub-Saharan Africa, Central Asia — economies with limited white-collar automation exposure, though this may simply reflect delayed adoption rather than structural immunity.
The EU has accelerated implementation of the AI Act to include labor market protections — requiring companies that deploy AI systems to conduct "social impact assessments" and provide 6 months' notice before automating positions. Critics argue the regulation will slow European AI adoption while doing little to protect workers.
The Biden-era AI executive order has been partially rolled back by Trump, but the administration announced a study on creating a "transition fund" financed by a tax on AI company revenues. No legislation has been introduced, and the study is not expected to produce recommendations until late 2027.
China has not publicly acknowledged AI-driven job displacement, instead framing AI as a tool for "upgrading" the workforce. Behind the scenes, Chinese tech companies are automating as aggressively as their American counterparts.
Brazil's Congress has begun preliminary discussions on AI labor regulation but has not advanced past committee stage. The lack of urgency reflects Brazil's lower immediate exposure — but the BPO sector in cities like São Paulo and Curitiba is growing, and with it, vulnerability.
Goldman Sachs projects that AI automation will displace 2.5-4 million jobs globally by the end of 2027, while creating 1.5-2.5 million new positions — a net negative of approximately 1-1.5 million jobs. But the bank emphasizes that these aggregate figures obscure the human reality: each displaced worker faces months of uncertainty, financial pressure, and the challenge of reinventing a career in a labor market that is changing faster than institutions can adapt.
The report concludes with a warning that history will judge the current moment not by the technology that was built, but by the social infrastructure that was — or was not — constructed to absorb its impact.
Granular sector data within the April 2026 Goldman Sachs macroeconomic assessment demonstrates that the displacement of 300,000 positions in ninety days reflects structural task compression rather than cyclical recessionary layoffs. Unlike previous automation waves that mechanized assembly-line manufacturing while expanding administrative payrolls, multimodal autonomous agents directly absorb routine cognitive workflows in legal document discovery, financial compliance auditing, tier-one software debugging, and multilingual customer operations.
For monetary policymakers at the Federal Reserve and the European Central Bank, this decoupling between surging corporate operating margins and contracting entry-level white-collar hiring complicates traditional Phillips Curve inflation modeling, accelerating policy debates over sovereign compute taxes, portable retraining credits, and corporate AI transition disclosures.
Sources: Goldman Sachs Global Investment Research — AI Employment Impact Report Q1 2026, Reuters, Bloomberg, McKinsey Global Institute, World Economic Forum Future of Jobs Report 2026
