The Trillion-Dollar Disconnect
The global textile and apparel market, valued at US$535 billion to US$1.11 trillion in 2024, employs approximately 430 million people representing 11.9% of the global workforce. This makes it one of humanity’s largest employment sectors, with women forming 75-80% of the workforce in garment manufacturing.
Yet these industries face a profound contradiction. While PwC estimates AI could contribute US$15.7 trillion to the global economy by 2030, and McKinsey projects US$13 trillion in additional output (1.2% annual GDP boost), very few organizations are realizing these transformative benefits today. Generative AI alone could add US$2.6 to US$4.4 trillion annually, boosting labor productivity growth by 0.1% to 0.6% per year through 2040.
Despite this euphoria, sophisticated investors remain skeptical. Long-term bond yields have fallen during major AI releases, suggesting capital markets are unconvinced AI will generate sustained growth. For labor-intensive sectors like apparel—where the Asia-Pacific region dominates with 48.7% market share—this paradox is not theoretical debate but existential challenge.
The ‘Process-Forward’ Trap
Most companies take a ‘process-forward’ approach, layering AI onto existing workflows while leaving fundamental operating models untouched. In manufacturing hubs where textile exports rose 7% to US$21.4 billion in recent periods, companies use AI tools to automate inventory sorting, but supply chain architecture remains unchanged.
AI is not merely an efficiency tool; it redefines business operations. Leaders must adopt a future-back approach, designing functions from scratch for an AI world. This means reimagining supply chains not as linear ‘design-source-manufacture-sell’ sequences, but as integrated, AI-driven ecosystems moving from forecasting to granular prediction, and from mass production to AI-enabled mass-customization.
Productivity, Displacement, and the Endangered Consumer
Firm-level data reveals complex patterns. In customer service, generative AI boosts productivity of novice workers, narrowing gaps with high-skill counterparts. In materials science, it doubles output of top researchers while leaving bottom-tier scientists unaffected.
This dichotomy reshapes the sector asymmetrically. In design, AI amplifies top-tier talent, widening gaps. In logistics, it assists novice managers, narrowing competency differences.
The greater concern is widespread displacement. The World Economic Forum estimates AI could eliminate 92 million existing roles by 2030, with McKinsey suggesting 375 million people may need to change jobs or acquire new skills. Analyses project 80% of the U.S. workforce could see at least 10% of their tasks affected, with 30% of tasks potentially automated by 2030.
In manufacturing, AI-driven HR automation carries ethical risks perpetuating biases in recruitment data. In sectors where workers receive only 0.6% of a standard t-shirt’s cost, and informal employment reaches 72% in some regions, such automation could exacerbate existing vulnerabilities.
This links to bond market skepticism. Investors fear AI will produce disruption without growth, with benefits flowing to tech shareholders while costs job losses and wage pressure, spread across populations. For an industry where global consumer spending totals US$2.4 trillion, projected to reach US$2.88 trillion by 2029, a future of concentrated wealth but lower aggregate consumption is catastrophic.
The New Bottleneck: From Labor Costs to Kilowatt Costs
The AI race has become a contest of kilowatts. The International Energy Agency projects global electricity consumption from data centers will more than double to 945 terawatt-hours by 2030, equivalent to Japan’s total consumption with AI as the most significant driver. Data centers will rise from 1.5% of global electricity consumption in 2024 to 3% by 2030. Goldman Sachs estimates data center power demand will grow 160% by 2030, reaching 8% of US power compared with 3% in 2022.
China is positioning to win by creating an “energy-compute flywheel.” China’s installed capacity increased 14.6% in 2024 to 3,348 GW, with solar capacity surging 45.2% (+277 GW) to 887 GW, and wind power increasing 18% (+80 GW) to 521 GW. In 2024 alone, China added 429 GW of new capacity, with wind and solar accounting for 83% at 356.5 GW. This fulfilled the 1.2-terawatt renewable target originally set for 2030—five years ahead of schedule.
China simultaneously champions open-source AI models with inference costs one-tenth of proprietary Western models. More clean power enables more cheap compute, which optimizes energy grids.
This presents dire warnings to manufacturing hubs. Future competitiveness will no longer be decided solely by wage bills, but by access to cheap, stable, clean energy to power AI-driven operations. The IEA estimates up to 20% of planned data center projects could face delays without significant transmission infrastructure investments. Western nations face constraints from aging grids and slow permitting. For developing nations, the challenge is exponentially greater.
The Capital Squeeze and Financial Contagion Risk
This transformation requires immense capital, creating systemic risks fundamentally different from past technology booms. Unlike the 2000 dot-com bust, today’s AI boom is fueled by debt.
Hedge fund leverage has reached historical highs, with the top 10 funds accounting for 40% of total repo borrowing with leverage ratios of 18:1 as of Q3 2024. Bank credit commitments to non-bank financial institutions grew to US$2.3 trillion in Q4 2024. These funds heavily finance the AI boom, including data center construction, with concentration in tech stocks.
A market crash triggered by AI’s failure to meet expectations could cause a chain reaction of margin calls and forced deleveraging, precipitating financial crisis. While AI-driven solutions could reduce global carbon emissions by 10% by 2030, sophisticated AI-driven cyberattacks could cost businesses US$10 trillion annually by 2030.
For developing countries where apparel and textile manufacturing forms economic cornerstones, this creates a perilous squeeze. These nations must invest in AI and energy infrastructure while facing declining foreign direct investment and geopolitical vulnerability. They risk becoming dependent on private AI providers from Silicon Valley and Shenzhen.
To avoid this, nations must prioritize supportive regulatory environments, attracting FDI, and most critically, investing in human capital. The 21st-century growth sectors require well-educated workforces even for upstream opportunities.
The Path Forward
We are navigating this transformation while flying blind, failing to measure AI’s true economic impact. Traditional metrics like GDP are lagging indicators; transformative technologies took decades to appear in official statistics. We lack systematic data on which firms use AI, in which sectors, and for what purposes.
With 82% of shoppers aged 26-35 having purchased clothes online, yet 58% preferring in-store shopping, and 92 million tons of textiles discarded annually, the industry faces sustainability pressures alongside technological transformation. Half of today’s work activities could be automated between 2030 and 2060, with a midpoint in 2045, roughly a decade earlier than previous estimates.
For global apparel and textile sectors, success will not come from incremental pilots. It requires the difficult, multi-year journey of building new future-back operating models powered as much by kilowatts as by human creativity.
This transformation must balance three imperatives– i) maintaining livelihoods of hundreds of millions of workers, ii) meeting sustainability demands, and iii) capturing productivity benefits AI promises. The industry that clothed humanity through the Industrial Revolution must now reinvent itself for the Intelligence Revolution or risk obsolescence in the very transformation it helped finance through consumer spending.
The question is no longer whether AI will reshape these industries, but whether the transformation will create shared prosperity or concentrated disruption. The data suggests we are precariously balanced between both futures.
Author:
Enamul Hafiz Latifee is a Trade and Development Economist, Lifetime Member of Bangladesh Economic Association (BEA), and Founder of EHL Market Intel.
