Economic Survey flags rising risks for startups in technology and AI

The report backs innovation as a growth driver but warns that costly capital, long-duration investments and narrow margins of safety could test India's early-stage enterprises

India’s innovation and entrepreneurial activity is unfolding amid a global environment marked by fragility, leverage and narrowing margins of safety. The Economic Survey 2025–26 acknowledges the role of technology and innovation in sustaining growth, but cautions that recent phases of highly leveraged technology and AI-infrastructure investment rest on optimistic execution timelines, narrow customer concentration and long-duration capital commitments.

 

“Fragility, uncertainty and episodic shocks are increasingly structural features of the system,” according to the Survey, with countries facing “a narrower margin of safety.”

 

With capital remaining structurally expensive and external buffers finite, growth that is detached from productivity gains, export competitiveness and balance-sheet strength carries rising risks. In this environment, endurance, execution discipline and competitiveness — rather than exuberance — will determine which enterprises endure. For startups, the most immediate implication lies in capital.

 

 

The Survey makes an assessment of why money remains expensive in India, arguing that “India’s high cost of capital is a structural macroeconomic outcome” tied to persistent current account deficits and dependence on foreign savings. In practical terms, this suggests that the era of abundant, low-cost funding — particularly for long-gestation, loss-making ventures — is unlikely to return soon.

 

Artificial Intelligence

 

The Survey turns to technology and artificial intelligence. Referring to global trends, it cautions that the recent phase of highly leveraged AI-infrastructure investment has exposed business models that are dependent on optimistic execution timelines, narrow customer concentration and long-duration capital commitments. While the comment is global in scope, its relevance for India’s AI and deep-tech startups is clear: leverage, scale assumptions and delayed monetisation now carry systemic risk.

 

It said a correction in this segment would not end technological adoption, but it could tighten financial conditions, trigger risk aversion and spill over into broader capital markets. If such developments were to coincide with geopolitical escalation or trade disruption, the resulting interaction could produce a sharper contraction in liquidity, a sudden weakening of capital flows, and a shift towards defensive economic responses across regions. “While this remains a lower-probability scenario, its consequences would be significantly asymmetric. The macroeconomic consequences could be worse than those of the 2008 global financial crisis,” the Survey said.

 

At the same time, the Survey does not dismiss innovation. A full chapter on artificial intelligence positions AI as economically consequential, while also highlighting governance, data architecture and safety. The emphasis, however, is on phased adoption and institutional readiness rather than unrestrained experimentation. In this environment, execution discipline matters as much as technological ambition.

 

The Survey said startups focused on curating AI training data have yet to scale in India, limiting the country’s ability to tap its vast data resources. India represents about 2 per cent of global startups in AI training data, compared with 40 per cent in the US, 21 per cent in the European Union and 9 per cent in the UK, while China accounts for roughly 5 per cent, the Survey said, citing World Bank data.

 

“India holds considerable potential advantage in terms of domestic data sources, but this asset remains underutilised,” the Survey said.

 

The report says global AI development is diverging between capital-intensive frontier models and decentralised, application-driven approaches, arguing India should focus on smaller, sector-specific models given constraints on computing power and funding.

 

Where the Survey appears most constructive for startups is in manufacturing, logistics and export-linked activity. It repeatedly stresses that services exports, while valuable, are “not a substitute for the goods-based export ecosystems that ultimately underpin durable external and currency stability.” For startups, this points towards opportunities in industrial technology, supply-chain software, energy efficiency, electronics, and platforms that help small manufacturers scale.

 

The role of the state also features prominently. Drawing on the idea of an “entrepreneurial state”, the Survey argues for mission-mode platforms, procurement-led innovation and deregulation that shifts the government from controller to enabler. These signals matter for startups operating in sectors such as climate, defence, health and infrastructure, where the state is often the largest customer.