In today’s digital order, a silent form of neo-colonialism is emerging – not through armies or empires, but through algorithms. Corporations in the Global North extract data from the Global South – under the guise of “free” connectivity and convenience.  Meta’s “Free Basics” project promised “internet for all” for millions in India and across Africa, but fenced users within Meta’s curated ecosystem. Google’s AndroidOS, powering billions of smartphones across Asia, Africa, Latin America, quietly transmits extensive user data – including device identifiers, app usage, and location metadata – even when not actively used, to the Northern servers – 20 times more than  Apple’s iPhones – according to Trinity College Dublin . Similarly, Amazon Web Services (AWS) store critical financial and government data from developing countries on servers situated in Global North. This reflects a form of modern exploitation that parallels historical extraction practices, replacing cotton fields and gold mines with clicks, scrolls, and selfies, while igniting fierce debates over data sovereignty, epistemic justice, and digital enclosure. Scholars Nick Couldry and Ulises Mejias call this “data colonialism” – a modern extraction regime where human experience becomes the raw material for profit. The Global South, once carved up for its natural resources, now serves as a plantation of information feeding Silicon Valley’s capital machine.

Unseen Workers in the Digital Factory

Beneath the glamour of “free” apps lies an invisible labour force. Every scroll, search and post adds value to Big Tech’s empire, yet the users – ‘digital labourers’ receive nothing in return. As Tiziana Terranova (2000) observed, digital labour encompasses all  unremunerated activities that produce online culture and information. Today, it also includes underpaid gigwork that props us to the attention economy.

An average user spends over two hours and twenty minutes daily on social media through various activities – such as posting on social media, querying search engines, or navigating ride-hailing apps – thus fueling more than 15 billion collective hours of engagement each day (DataReportal, 2024). Meta’s 2024 revenue soared to $164.5 billion – almost entirely from advertising (Macrotrends, 2024). Alphabet (Google) reported $307 billion in total revenue, with $237 billion derived only from targeted ads the same year (Macrotrends, 2024; Alphabet Annual Report, 2024). The global digital advertising market surpassed $600 billion in 2024 and is projected to reach $870 billion by 2027 (IAB, 2025), illustrating the immense profitability of algorithmic attention economies. Strategic acquisitions like Meta’s $19 billion purchase of WhatsApp and Google’s $1.65 billion acquisition of YouTube have consolidated control over major communication channels and content creation. The modern internet users do not merely participate in the digital economy, rather labour within this digital economy, turning every scroll, selfie, and status update into profit streams that benefit corporations far away. This digital labour echoes the resource extraction of past industrial eras—only now, data replaces coal and oil. Every like, share, or geotag feeds AI systems, fine-tunes ads, and fuels corporate profit, concentrating digital wealth in the hands of a few. Attention is the new oil, enriching Silicon Valley titans – Mark Zuckerberg, Elon Musk, Jeff Larry Page, and Sergey Brin – whose combined wealth now exceeds $900 billion, eclipsing the GDP of many Global South nations. By  2025, Zuckerberg’s net worth alone hovers around $160 billion, while Musk exceeds $220 billion (Forbes, 2025), fueled by his platforms (X, formerly Twitter; Tesla’s data-driven autopilot systems; and SpaceX’s satellite networks) that depend on behavioral and geospatial data. 

Meanwhile, the digital underclass toils unseen.  In Kenya and Uganda, low-wage workers perform hidden tasks that sustain artificial intelligence systems such as labelling images, annotating text, and moderating content, often at a significant psychological cost. A Time investigation revealed that in Kenya, content moderators employed by outsourcing firms like Sama, earned as little as $1.32 an hour for cleaning toxic data for OpenAI, leaving many traumatised by constant exposure to graphic material. Across the Global South, millions of “ghost workers” label images, tag text and train AI systems, enduring long hours filtering harmful content to keep online spaces “safe” for consumers in the Global North. The World Economic Forum (2025), estimates that over 60% of Southern workers – roughly two billion people – exist in informal, unprotected digital economies.

Data Colonialism and the Algorithmic Empire

Big Tech’s dominance mirrors earlier colonial economies: systematic extraction, accumulation, and commodification of cultural and behavioural data. Massive repositories like Common Crawl and LAION, feed large-scale models such as ChatGPT and Stable Diffusion, absorbing global knowledge without consent or compensation. Simultaneously, embedded analytics in applications and cloud infrastructures convert everyday interactions like – digital gestures and messages, location traces, and clicks, into streams of commodified data. Acquisitions such as Meta’s takeover of WhatsApp and Google’s dominance over Android further consolidate these extraction capabilities, ensuring that informational capital continues to flow Northward.

This neo-extractive dynamic has begun to reshape law, labour, and global governance. In early 2023, Getty Images filed a lawsuit against Stability AI claiming its Stable Diffusion model was trained on over 12 million copyrighted photographs without authorization, potentially resulting in $1.8 trillion in theoretical damages (Ars Technica, 2023). Similarly, The New York Times and several authors have sued OpenAI and Microsoft for using copyrighted material and books to train ChatGPT. These cases illuminate the structural asymmetry of the digital economy: Global South produces data; while Global North monopolize computation, patents, and profits.

Some resistance is noticeable in recent times. India’s Digital Personal Data Protection Act (2023) and Brazil’s  Lei Geral de Proteção de Dados, (2018), signaling growing resistance against such global imbalances and promote data sovereignty. However, despite these efforts, major tech companies like Google, Meta, and Amazon – lobbying over $70 million in 2024 alone (Statista, 2025) – continue to weaken enforcement and preserve extractive pipelines. 

This digital order also reproduces social hierarchies. UNESCO’s 2024 study found AI associating women with domestic roles four times more frequently than men, while linking male names to “career” and “business” categories, reflecting entrenched gender bias in algorithmic outputs (UNESCO, 2024). Research on large language models and AI-generated images likewise reveals skewed representation: in a sample of 15,300 AI-generated occupational images, women were underrepresented in technical and leadership roles and depicted with more submissive facial expressions (Cho et al., 2023). Artists from Africa and Latin America, report that their creative works which include visual motifs, folklore, language among others, are used in generative outputs without their permission or compensation, echoing earlier colonial appropriation of indigenous knowledge.  What was once the extraction of rubber, gold, or cotton now manifests as the extraction of cognition, culture, and creativity. This may be called neo-extractivism, where cognition, culture and creativity replace land and minerals as the colonised frontier. In this algorithmic plantation economy, the commons of human expression are fenced off as proprietary data assets, reinforcing a digital dependency in which the Global North owns the means of cultural and computational production.

Reclaiming the Digital Future

Through a Marxist lens, the digital economy becomes a factory of alienation – where users are reduced to data points, labourers outsourced to invisibility, and profits extracted at planetary scale.  The surplus value that Marx once located in factories now hides in algorithms. To dismantle this new colonial matrix, the Global South must reclaim data sovereignty. Equitable governance, participatory policymaking, and cross-regional cooperation can transform digital dependency into autonomy. Only by asserting ownership over data and algorithms can the South escape serving as the North’s digital underclass. The path forward demands digital decolonization—a reimagining of technology not as an instrument of extraction, but as a platform for justice, inclusion, and shared prosperity. The struggle for sovereignty has moved from the soil to the server.

(Dr. Manasi Sinha (Ph.D, JNU) is assistant professor, Department of Political Science, Easwari School of Liberal Arts, SRM University, Andhra Pradesh, India. She is a Erasmus Mundas Fellow,  University of Warsaw, Poland. Views expressed are personal. She can be reached at manasi.s@srmap.edu.in. Her co-author is a student at School of Science and Engineering, SRM University.)