AI or die: What India's public sector must do next
"Embrace AI or die." That was Jack Hidary's blunt message at the World Economic Forum Annual Meeting. The CEO of SandboxAQ believes agencies and enterprises that adopt AI will grow; those that delay will fade. India, he says, is positioned to benefit if it moves with intent.
His focus isn't content tools. It's AI plus quantum methods applied to the physical economy-healthcare, energy, materials, navigation, and cybersecurity. That's where India's stakes are highest.
Why this matters for government
Nearly 80% of India's economy runs on physical systems-railways, power, telecom, manufacturing, and infrastructure. AI can compress timelines, cut waste, and improve reliability across these. The upside is national-scale productivity. The risk is falling behind and exposing critical infrastructure.
Where AI delivers immediate public value
- Public health and pharma: AI models can narrow drug candidates for cancer, Alzheimer's, and infectious diseases, shrinking discovery cycles that often stretch 15 years.
- Materials and infrastructure: Simulate new compounds for stronger rails, resilient concrete, and lighter transport components.
- Energy and industry: AI-guided catalysts to convert hydrocarbons into cleaner products with better yield and lower cost.
- Navigation and logistics: Better routing, fewer delays, and smarter maintenance for rail and freight.
- Cybersecurity and post-quantum readiness: Detect threats faster, automate response, and migrate to cryptography that can withstand future quantum attacks.
SandboxAQ, spun out of Alphabet in 2022, works with banks, companies, and governments globally, including in the US. Its backers include Eric Schmidt, Marc Benioff, and T Rowe Price-signals that capital is already moving to this space.
The security wake-up call
Cyber risk is immediate. Critical targets include banking, telecom, and public utilities. India's major IT service providers-Infosys, Wipro, TCS-carry vast global data footprints that raise the stakes.
Advanced attacks require advanced defenses. AI-driven detection, cryptographic inventory, and a rapid move to post-quantum cryptography (PQC) must land on every CIO's plan. For context, see WEF's analysis of cyber threats and resilience priorities here.
India's IP opportunity
Hidary's point is simple: build intellectual property at home. Pharma is the clearest path-AI-driven molecular design can shift India from being primarily a manufacturer to a generator of new medicines and platform technologies.
The same applies to catalysts, sensors, and advanced materials. Public funding plus industry co-development can turn national challenges into exportable IP.
Policy and procurement moves to act on now
- Agency AI roadmaps: Within each ministry, pick 3 use cases with clear ROI-health, mobility, and security. Set targets, owners, and timelines.
- Cyber modernization: Inventory cryptography across government systems. Begin PQC pilots in banking, identity, and telecom. Require vendors to provide PQC migration plans.
- Data pipelines: Stand up secure datasets for health, materials, and energy. Standardize access with privacy and audit controls.
- Standards and risk: Adopt the NIST AI Risk Management Framework and ISO guidance for AI assurance. Tie compliance to procurement.
- Public-private build programs: Co-fund pilots with industry in drug discovery, industrial catalysts, and infrastructure materials. Share results and best practices across states.
- Talent and upskilling: Train civil servants, CISOs, and program managers on AI literacy, evaluation, and vendor management. Require AI capability in key roles.
- Compute and cloud: Pre-negotiate GPU and cloud capacity. Ensure data residency, encryption, and incident response terms are non-negotiable.
If you need structured learning paths by role, explore focused options here.
Practical 30/90/180-day plan
- 30 days: Name an AI lead in each ministry. Identify top 3 use cases and critical crypto dependencies. Set success metrics.
- 90 days: Launch pilots in one hospital network, one state utility, and one transport corridor. Begin PQC pilot in a high-risk system.
- 180 days: Scale what works. Bake AI and PQC requirements into procurement. Publish a cross-ministry scorecard and budget for year two.
The bottom line
"AI or die" isn't hype-it's a selection event. Governments that move first reduce risk, lower costs, and set standards industry will follow. India has the scale and the sectors; now it needs speed, clear ownership, and disciplined execution.
As Hidary notes, AI won't stay a "nice to have." It decides which systems grow-and which get left behind.
Your membership also unlocks:
