The Digital India Act’s AI provisions take effect April 1, 2026. If you’re building with AI, deploying AI systems, or using AI for customer interactions, there are compliance requirements you need to meet in the next 30 days.
We’ve distilled the 847-page framework into what actually matters for startups.
Who This Applies To
Any company that:
- Deploys AI systems that interact with Indian consumers
- Uses AI for automated decision-making affecting users
- Processes personal data using AI/ML systems
- Offers AI-as-a-Service to Indian businesses
The 5 Key Requirements
1. AI System Registration
Deadline: April 1, 2026
All “high-risk” AI systems must be registered with MeitY. High-risk includes: credit scoring, hiring decisions, healthcare diagnostics, and customer service automation.
2. Transparency Disclosures
Deadline: April 1, 2026
Users must be informed when they’re interacting with AI. This affects chatbots, voice agents, and automated email systems.
3. Algorithmic Impact Assessments
Deadline: June 1, 2026
Companies with 1M+ users must conduct and publish algorithmic impact assessments for AI systems affecting user outcomes.
4. Data Localization for AI Training
Deadline: October 1, 2026
AI systems trained on Indian user data must maintain training data copies within India.
5. Human Override Requirements
Deadline: April 1, 2026
Automated decision systems must provide human appeal mechanisms for decisions with “significant impact.”
Penalties
Non-compliance penalties range from ₹10 lakh to ₹500 crore depending on company size and violation severity. The framework includes a 90-day remediation window for first-time violations.
What You Should Do This Week
- Audit all AI systems in your stack
- Identify which fall under “high-risk” categories
- Prepare registration documentation
- Update user-facing disclosures
- Establish human review processes