#5: India Launches SAHI Healthcare Strategy, UNDRR Demands Legal Frameworks for AI in Disasters, UNICEF Mandates Child Rights Assessments
AI for Impact Daily Briefing — February 17, 2026
🔥 Top Stories
1. India Launches SAHI & BODH — First National AI Healthcare Strategy in South-East Asia
India's Health Minister JP Nadda launched SAHI (Strategy for Artificial Intelligence in Healthcare for India) and BODH (Benchmarking Open Data Platform for Health AI) at the India AI Impact Summit. WHO called India "the first country in South-East Asia to adopt a comprehensive national AI strategy for health." AI is already integrated for TB risk prediction, diabetic retinopathy screening, and telemedicine. BODH creates a quality gate ensuring AI tools used by frontline health workers are safe and benchmarked against real-world outcomes.
2. India AI Impact Summit Opens — Largest Global AI Governance Gathering Yet
The India AI Impact Summit 2026 opened at Bharat Mandapam, New Delhi — the first global AI safety summit hosted in the Global South. PM Modi, Macron, Lula, UN Secretary-General Guterres, and tech CEOs are attending the 5-day event. Released the "AI Compendium" playbook and "Seven Chakras" framework for international AI cooperation. 250,000+ visitors, 600+ AI startups, $68 billion in investment commitments through 2030.
3. UNDRR Calls for Legal Framework for AI in Disaster Management
At the India AI Summit, UNDRR's Sujit Mohanty declared that AI in disaster management urgently needs legal and policy frameworks. Ground-level officials like district magistrates cannot rely on AI-driven tools without legal backing for their decisions. India's AI disaster prediction work offers a blueprint that could reduce $200 billion in global annual disaster losses — but only if governance keeps pace with the technology.
4. UNICEF Releases "Guidance on AI and Children 3.0" — Mandates Child Rights Impact Assessments
UNICEF's most actionable guidance yet calls on governments to mandate Child Rights Impact Assessments (CRIA) for any AI system in public services. Ten core requirements including "Safety by Design," prohibitions on biometric data harvesting from children in refugee settings, and algorithmic audits for fairness across linguistic groups. AI must be "safe by default."
5. Salesforce/Good360 AI Agent Transforms Disaster Relief Matching
Good360 used Agentforce AI agents during the 2025 LA wildfires. Product-to-nonprofit matching time dropped from 30 minutes to 10 minutes. The AI anticipated needs humans missed — locating tents and proactively suggesting workboots. The UN's "Early Warnings for All" initiative uses AI-powered NLP to translate disaster alerts into multiple languages.
6. IRC Airbel Impact Lab: AI-Powered ReachMap Finds 10-20% More Unvaccinated Children
Amid USAID funding cuts (83% of grants cancelled), the IRC's ReachMap combines satellite imagery and AI route planning to reach 10-20% more zero-dose children while cutting costs 15-25% per child. "Follow the Forecast" uses predictive data for anticipatory cash transfers before disasters strike. Critical tools as humanitarian organizations face the worst funding crisis in a generation.
7. Everdrone Deploys Autonomous Emergency Drones in Stockholm
Everdrone signed a multi-year agreement with Region Stockholm to deploy autonomous drones delivering defibrillators for cardiac arrest emergencies. Drones launch simultaneously with ambulances, using computer vision for obstacle avoidance and BVLOS navigation. Live video to dispatchers for scene assessment. Pilot to permanent municipal integration.
8. University of Twente: Drones Cut Humanitarian Delivery Delays by 85%
Research across six historical disasters (Haiti earthquake, Indonesia tsunami, Hurricane Dorian) found AI-optimized drone logistics reduce transport costs 7-16%, increase reachable locations 11-56%, and decrease delivery delays up to 85%. Most effective for the hardest-to-reach 10-20% of locations in the critical first 72 hours.
📅 Upcoming Events
India AI Impact Summit 2026
- Dates: February 16–20, 2026
- Location: New Delhi, India
- Why it matters: Largest global AI governance gathering. Tracks on AI for Social Good, Healthcare, Disaster Resilience. Currently underway.
NVIDIA GTC 2026
- Dates: March 16–19, 2026
- Location: San Jose, CA + Virtual
- Why it matters: Major tracks on climate modeling, healthcare AI, and public sector deployment.
AAAI Spring Symposium: AI + HADR
- Dates: April 7–9, 2026
- Location: Burlingame, CA
- Why it matters: Premier AI + disaster relief academic venue. Speakers from ICRC, Harvard/Google.
- Link: hadr.ai
Responsible AI Forum 2026
- Dates: May 19–20, 2026
- Location: Munich, Germany
- Why it matters: Research, policy, and practice on ethical AI deployment.
- Link: responsibleaiforum.com
🌍 Active Disaster Monitoring (GDACS/OCHA)
- M6.0 earthquake — Russia: February 16, depth 43km, ~5,000 exposed at MMI IV intensity. Monitoring ongoing.
- M5.7 earthquake — Solomon Islands: February 16. Pacific ring of fire activity.
- Flood alerts: Active across United States, Peru, Turkey, Afghanistan, Ukraine, Montenegro.
- Sudan: War approaching fourth year. Humanitarian agencies appeal for refugee support as displacement accelerates.
- Yemen: Rapid displacement tracking period active (February 8-14).
💡 Signal Watch
- USAID Funding Crisis: 83% of grants cancelled. UK and Germany following with cuts. IRC reports 6M+ clients with service interruptions, 2M lost services entirely.
- Google Flood Hub Expansion: Now covering 80+ countries, 460M+ people with 7-day forecasts. Triggering anticipatory cash transfers in Bangladesh and Nigeria.
- Ethical Counter-Narrative: New academic research on dual-use risks of satellite/CV technologies in conflict zones.
Sources: Al Jazeera, BBC, CNBC, NDTV, The Hindu, Tribune India, Business Standard, Devdiscourse, Hindustan Times, UNICEF, Deccan Herald, Salesforce, IRC, fundsforNGOs, University of Twente, Unmanned Systems Technology, Everdrone, GDACS, OCHA, ReliefWeb