AI for Good Isn’t Just a Buzzword: Lessons from the Field
- Caroline Van Cauwelaert
- Jun 25
- 2 min read
Updated: Jun 27
In recent years, "AI for Good" has become a fixture on global conference panels, corporate CSR decks, and international development agendas. But for us at EPCON, it’s not a slogan — it’s a daily commitment to making artificial intelligence useful, ethical, and impactful in some of the world’s most underserved health systems.

We work in places where data is sparse, infrastructure is stretched, and lives depend on better decisions being made faster. And from Nigeria to the Philippines, from Central African Republic to Uganda, we’ve learned a few important things about what it takes to turn AI into real public health impact.
1. No One Needs a Shiny Dashboard That Doesn’t Work Locally
AI solutions built in isolation — without context, without local engagement — often fail. That’s why EPCON partners directly with Ministries of Health, NGOs, and regional teams. We don’t just hand over a tool. We co-design solutions that align with existing workflows and realities, human resources, and routine data.
2. "No Data" Is Not the Same as "No Insight"
In many countries, traditional surveillance systems work in silos and often show gaps in data completeness and accuracy. That doesn't mean we can't act. EPCON's AI model uses evidence from other parts in the country and proxy indicators, like mobility, poverty, access to care, weather — to estimate disease risk at a neighborhood level, even in places with no reported cases. This allows Ministries to prioritize where to send diagnostic tools, even before the first case is confirmed.
3. Transparency Builds Trust
We believe AI must be explainable — especially in public health. Our models don't just spit out a risk score; they show which factors contributed to the risk. This helps health officials make informed, accountable decisions — and helps communities understand why action is needed in their area.
4. Local Capacity is the Endgame
Every project includes knowledge transfer. We train regional teams to support them interpret the insights from our Epi-control platform and integrate these into planning cycles. AI should never be a foreign language — it should be a tool people understand and use with confidence.
Looking Ahead
The promise of AI in global health isn’t an abstract potential — it’s tangible outcomes: fewer missed diagnoses, faster responses to outbreaks, and smarter use of scarce resources.
At EPCON, we keep building tools that work not just in theory, but in reality. Our goal is simple: make AI a integral part of health systems, not an accessory.