Industry experts are urging India to resist the temptation to import a ready-made artificial intelligence safety model from the United States and instead build a framework shaped by local realities, including linguistic diversity, large-scale deployment and an increasingly application-driven startup ecosystem.
The debate has sharpened after a voluntary AI safety pact was signed in the US last week by President Donald Trump and top technology firms, a move that has drawn global attention to how governments are trying to manage the risks of fast-moving generative AI systems. But experts said India should not treat that pact as a template. They warned that a direct copy would overlook the country's distinct market structure, where AI is being embedded not only in consumer products but also in enterprise software, fintech, healthcare, education and government-facing services.
Local Risks, Local Rules
India's AI challenge is not identical to that of the US or Europe. The country's digital ecosystem must operate across dozens of major languages and dialects, often with uneven data quality and limited standardisation. That creates a different safety problem: models can behave unpredictably in underrepresented languages, produce biased or inaccurate outputs, and amplify misinformation at scale. Experts said any Indian framework must therefore address testing, auditing and accountability in multilingual contexts rather than rely on broad principles alone.
They also noted that India's AI adoption is increasingly application-led, with startups and large enterprises racing to deploy tools that automate customer support, credit assessment, content generation and workflow management. In such an environment, safety cannot be treated only as a frontier-model issue. It must extend to how AI is integrated into products, how outputs are monitored, and who is responsible when systems fail.
A key concern is that a voluntary pact, while politically useful, may not provide the enforceable guardrails India needs. Industry voices said India should instead develop a framework that combines technical standards, sector-specific rules and clear disclosure obligations. That could include model evaluation benchmarks, red-teaming requirements, incident reporting norms and transparency around synthetic content.
Startup Growth, Guardrails
For startups, the central question is whether regulation will slow innovation or create a more credible market. Experts argued that well-designed safety rules can do both: reduce reputational risk for founders and increase trust among customers, investors and regulators. In a market where many AI products are still experimental, clarity on liability and compliance could help separate serious companies from opportunistic ones.
At the same time, they cautioned against a heavy-handed regime that would burden early-stage firms with costs they cannot absorb. India's framework, they said, should be proportionate, risk-based and phased, with lighter obligations for low-risk use cases and stricter scrutiny for systems used in finance, healthcare, education and public services.
The broader policy debate also reflects India's strategic position. The country wants to be seen not merely as a consumer of imported AI systems but as a global development and deployment hub. That ambition requires rules that are credible internationally but grounded domestically. Experts said India has an opportunity to shape a model that emphasises practical safety, inclusive language coverage and responsible deployment at scale.
What India Must Build
The consensus among industry observers is that India now needs a framework that is more than a statement of intent. It must define what safety means in the Indian context, identify which entities are accountable, and establish how compliance will be measured. Without that, experts warned, the country risks either over-regulating a still-maturing sector or under-regulating systems that are already influencing millions of users.
The policy direction is likely to matter most for startups and venture-backed companies, which are under pressure to move quickly while also proving that their products are trustworthy. As AI moves deeper into everyday business and public life, India's regulatory answer will shape not only innovation but also the credibility of the country's digital economy.