While Washington fixates on export controls and hardware restrictions, a quieter, more insidious dependency takes root in the foundational software layer.
For years, the global architecture of AI rested on an unquestioned orthodoxy; supremacy belonged to those who could marshal capital and computational brute force. Silicon Valley built its impregnable moat upon one assumption—that dominating the frontier required tens of thousands of GPUs and billions in infrastructure… a consensus that has now collapsed. The release of DeepSeek’s R1 and V3 models dismantled the premise that capital equals invincibility. By proving mathematical ingenuity and architectural efficiency can outperform raw power, it is clear that the geopolitical calculus of the AI arms race has shifted. This is no iterative update; it is a structural realignment of how sovereign computing power must be generated and distributed in the coming decade. The lesson for others—India included—is stark: silicon sovereignty is not bought with gold alone, but forged through algorithmic elegance. In this new era, intellectual agility outweighs brute force… the moat has evaporated, leaving only the raw intellect to lead.
The financial asymmetries of this breakthrough reveal the fragility of any monopoly built on spending… a haze of vulnerability. DeepSeek trained V3 using roughly 2,000 Nvidia H800 GPUs at $5.6 million, while R1 required a mere $294,000 in computational hours. In contrast, western incumbents deploy clusters exceeding 10,000 processors, burning through budgets upwards of $100 million for comparable benchmarks. The market reaction—a staggering single-day erasure of over $600 billion from Nvidia’s capitalization—signalled that the paradigm had shifted. If algorithmic efficiency bypasses endless hardware scaling, the primary defensive barrier protecting technology giants ceases to exist. Competition has pivoted: no longer who stockpiles silicon, but who engineers the leanest, most effective algorithms.
While Washington fixates on export controls and hardware restrictions, a quieter, more insidious dependency takes root in the foundational software layer. Control over open-source development dictates global technological standards; so it be. China’s Qwen model family has systematically eclipsed Meta’s Llama as the most downloaded open-weight ecosystem on Hugging Face. By early 2026, Qwen surpassed one billion total downloads—averaging over a million daily installations and spawning more than 200,000 derivative models. Between August 2024 and August 2025, developers using Chinese open models captured 17.1% of all downloads, edging past American developers. By flooding the ecosystem with free-to-deploy models, a strategy is established where infrastructure dependency bypasses restrictions… a soft-power hegemony capturing nearly 30% of global token usage among next-generation coders.
In response to this multi-front challenge, the ideological boundary separating American free-market principles from state industrial policy has all but dissolved. The notion of “Fortress Silicon” now relies on deep state intervention… a paradox. By August 2025, the American government converted $8.9 billion in CHIPS Act grants into a 9.9% equity stake in Intel, acquiring over 433 million shares. By spring 2026, this nationalized holding surged to approximately $35.4 billion. When the architect of the global free market essentially nationalizes its semiconductor champion for national priorities, it inadvertently validates state-directed models. This pivot to state capitalism in Washington complicates diplomatic efforts to rally allies against subsidies, as the United States now engages in identical market interventions.
Furthermore, the ambition of semiconductor reshoring harbours deep physical and economic contradictions… a labyrinth of dependencies. Decoupling relies on an interconnected supply chain that refuses to split along national borders. The US seeks to choke advanced logic chips for Chinese buyers, yet domestic innovation and new fabrication plants depend heavily on revenue from those very markets. In late 2025, the administration permitted Nvidia’s H200 sales via case-by-case licenses, prompting orders for over two million units. However, Beijing quickly suspended these imports, citing national security and the risk of relying on American technology. Critically, AI infrastructure remains tethered to Chinese manufacturing; China commands a 67% share of high-density interconnect boards. A targeted disruption here would instantly bottleneck global computational power… a policy trilemma where security, viability, and leadership cannot coexist.
Capital alone cannot forge silicon sovereignty without the human intellect and deep expertise required to operate advanced fabrication facilities. The American manufacturing renaissance collides with a severe deficit in specialized labour—a projected shortfall of 146,000 engineers and technicians by decade’s end. This crisis is exacerbated by funding pressures; recent legislative drafts held National Science Foundation budgets more than 40% below authorized levels. Conversely, sustained investment in human capital can cross a historic threshold… criticality reached. According to OECD data, China eclipsed the US in total R&D investment in 2024, deploying $1.03 trillion against America’s $1.01 trillion. This is a relentless volume game driven by long-term state planning; Chinese research expenditures expanded over 14% annually since 2004, double the American growth rate. Compounding advantages in patents and breakthroughs become structurally irreversible.
This critical mass of capital and talent is aggressively pushing AI into the physical economy… embodied intelligence. Humanoid robotics represents the inevitable next frontier of industrial automation; here, the gap widens rapidly. During 2025, Chinese manufacturers shipped up to 90% of the world’s humanoid robots. Companies like AgiBot and Unitree deployed thousands of units priced below $6,000, while western competitors measured shipments in the low hundreds. Hyper-integrated electronics supply chains along the Yangtze River Delta allow hardware prototyping to move from concept to reality in ten to fourteen days—dwarfing the twelve-week cycles common in western hubs. By shipping imperfect but functional machines into industrial settings, a loop is created to harvest proprietary physical data necessary to train next-generation spatial intelligence models.
Ultimately, true technological dominance lies not in laboratory benchmark scores, but in systemic industrial deployment… a flow of pure utility. A deliberate strategy embeds algorithms into traditional sectors to drive macroeconomic productivity. By early 2025, nearly three-quarters of China’s Fortune 500 equivalents adopted generative systems—achieving universal penetration in telecommunications and automotive manufacturing, plus 90% integration in financial services. Real-world implementations are sweeping; automotive giants now generate a third of software code via algorithms, while heavy manufacturers transformed operational data querying efficiencies. To sustain this ecosystem, Beijing formalized a $295 billion nationwide grid of state-operated data centres in 2026, mandating that 80% of technology be domestically sourced… transforming compute from a commodity into a regulated strategic asset.
The narrative framing this ascendance as a new “Sputnik moment” fundamentally misdiagnoses the reality of the modern global economy… it is a different kind of song. The mid-century Soviet space program was an isolated military endeavour in a closed system. Today, the AI revolution is inextricably bound to the world’s largest manufacturing base and embedded within western commercial supply chains. American sanctions did not arrest innovation; rather, they served as a brutal, inadvertent subsidy for algorithmic efficiency, forcing researchers to architect their way out of hardware starvation. As the global technology stack bifurcates into incompatible, parallel ecosystems, the rules of sovereign computing are permanently rewritten. The defining question is no longer whether one nation can constrain another… instead, nations must navigate a fractured landscape where absolute supremacy is replaced by inescapable interdependence. For observers like India, the revelation is clear: resilience emerges from constraints. When hardware vanishes, the intellect must soar… a dance of mind over matter.
*Brijesh Singh is a senior IPS officer and an author (@ brijeshbsingh on X). His latest book on ancient India, “The Cloud Chariot” (Penguin) is out on stands. Views are personal.
