Shaping inclusive economic growth in 2026

About the Center

(Header) About the Center

The digital transformation of European micro-businesses

Focus Areas

(Header) Focus Areas

Scaling impact together

Events

(Header) Events

Building trust in the age of artificial intelligence

Insights

(Header) Insights

(Footer) About the Center

(Footer) Focus Areas

(Footer) Events

(Footer) Insights

PostContent

In parts of rural India, the hot air hums with the sound of bees. The country is one of the world’s major producers of honey, with a rapidly growing market worth nearly $500B.Many beekeepers here are semi-nomadic, travelling cross-country to "chase the bloom", making sure their bees have access to nectar and pollen. The work is physically demanding, but the pressure is psychological, too. Beekeepers—often women—must manage volatile finances, plan investments, and carry the risk when a harvest fails.When it came to designing an AI-powered app to help these beekeepers improve yields and avoid losses, Monika Shukla, founder and CEO of <a href="https://data.org/our-work/challenges/artificial-intelligence-to-accelerate-inclusion-challenge/awardees/buzzworthy-ventures/" rel="noopener noreferrer" target="_blank">Buzzworthy Ventures</a>, realized that a standard approach wouldn’t work. For a start, women beekeepers often have limited and sporadic access to a smartphone, making it difficult for them to receive vital information around climate risks to their hives, for example.

InlineImage

InlineQuote

Monika Shukla

Founder and CEO, Buzzworthy Ventures

So she focused on designing <a href="https://thehumblebee.co/pages/for-beekeepers" rel="noopener noreferrer" target="_blank">BeeKind</a>, an AI-powered app aligned with those realities. By consulting extensively with the community, her team shifted from text-heavy menus to voice-first interactions in local dialects. They replaced technical jargon with the vernacular beekeepers use to describe a "patchy brood" or "restless bees." From the outset, this gave the community confidence that the app would work for them, allowing them to build trust in a technology that might otherwise feel daunting.The results validated the approach. BeeKind has been downloaded hundreds of times and is being swiftly adopted by those in the field. Recognized for its ability to turn advanced hive monitoring into accessible, intelligent advice for rural women, it also achieved success of a different kind. It was selected as one of <a href="https://www.mastercard.com/global/en/news-and-trends/press/2024/december/mastercard-center-for-inclusive-growth-and-data-org-announce-ai2ai-challenge-awardees.html" rel="noopener noreferrer" target="_blank">five winners of the inaugural Artificial Intelligence to Accelerate Inclusion (AI2AI) Challenge.</a>

Scaling inclusion responsiblyLaunched by the&nbsp;Mastercard Center for Inclusive Growth&nbsp;and&nbsp;data.org, <a href="https://www.mastercard.com/us/en/news-and-trends/press/2024/june/mastercard-center-for-inclusive-growth-and-data-org-launch-artificial-intelligence-to-accelerate-inclusion-challenge.html" rel="noopener noreferrer" target="_blank">the AI2AI Challenge</a> is a global call for AI solutions that advance inclusion and economic empowerment for underserved communities. The challenge drew more than 500 applicants from 82 countries, each proposing ways to harness the data and AI revolution for social impact.Shukla’s Buzzworthy Ventures, parent company to honey producers Humble Bee and creator of the BeeKind app, was one of five winners selected to receive grant funding, technical assistance, and mentorship. Collectively the Challenge winners, which also include <a href="https://www.mastercard.com/global/en/news-and-trends/stories/2025/ai-for-good-quipu-colombia.html" rel="noopener noreferrer" target="_blank">Quipu</a> from Colombia, <a href="https://www.idinsight.org/" rel="noopener noreferrer" target="_blank">IDinsight</a> and their project in Ethiopia, <a href="https://data.org/organizations/link-health/" rel="noopener noreferrer" target="_blank">Link Health</a> in the US and <a href="https://www.signpost.ngo/" rel="noopener noreferrer" target="_blank">International Rescue Committee’s Signpost</a>, have already benefitted more than 61,000 people and unlocked $3.4m in follow-on funding.What links the winners is not just their tech stacks, but their philosophy. Each demonstrates a mantra of using AI to augment human potential rather than replace it, treating trust as the essential infrastructure for making an impact at scale.

Perry Hewitt

Chief Strategy Officer, data.org

“Over the last year, our awardees have shown us what's possible when you apply AI to humanitarian challenges and underscored that meaningful solutions must anchor in local needs, build trust, and prioritize practical, scalable tools.”This is especially true at Buzzworthy, where the tech is in many ways secondary to the human connection.“One lesson we have learned is that AI adoption in rural contexts is less about algorithms and more about relationships,” says Shukla. “When communities feel seen and heard in the design, technology becomes a tool they trust, not something they tolerate. That’s the lens through which we continue to build BeeKind.”

Transparency builds confidence<a href="https://data.org/organizations/quipu/" rel="noopener noreferrer" target="_blank">Quipu</a>, another AI2AI Challenge winner, is helping to bridge a different type of trust gap—this time in Colombia. For millions of informal micro-entrepreneurs, accessing credit is a persistent struggle. Traditional banks rely on formal credit histories that these small business owners simply don’t have, leaving them invisible to the financial system.By using AI to build credit scores based on alternative data—things like phone usage patterns, business photos, and geolocation—Quipu is opening doors that would normally remain closed. But for this to work, users have to first trust an algorithm with their personal data.Mercedes Bidart, CEO and Co-founder of Quipu, argues that for this to work, she has to explain why she deserves that trust in the first place.

“Trust starts with who brings the message,” Bidart says. Quipu partners with brands the merchants already use daily, which act as credibility bridges. But the real breakthrough is in explainability. “We emphasize transparency: we explain what data we use, how it’s protected, and how it directly benefits them.”Quipu acts as a coach rather than a black box. Its interface provides nudges, telling a user that uploading more recent business photos or activating auto-debit could increase their credit limit. This shifts the dynamic from judgment to partnership.“The biggest risk is deepening the trust gap,” Bidart says. “If users feel like they’re being judged by an invisible, unknowable system, they’re less likely to engage. Opaque AI models can replicate biases or penalize informality, even when claiming to serve underserved markets. Without explainability, there’s no accountability.”Again, the proof is in the results. Quipu has seen changes in behavior. Merchants improve their digital footprints and repayment habits because they understand the benefits that will follow from this.

Efficiency with empathyWhile Quipu focuses on transparency,&nbsp;IDinsight&nbsp;in Ethiopia is proving that trustworthy AI must keep humans in the loop.Ethiopia’s health extension workers (HEWs) and community health workers (CHWs) are the backbone of the country’s primary care system, with around 40,000 workers providing health care for anywhere from 2,500 to 3,000 people each. IDinsight developed an AI-powered query tool designed to help these workers with maternal and child health questions.The tool retrieves accurate, approved medical guidelines instantly—but it’s designed strictly to support the health worker’s judgment, not to make the diagnosis for them.“The current version of the solution supports HEWs when they have tricky questions,” explains Sid Ravinutala, Chief Data Scientist at IDinsight. “They are unsure of the answer and they call a call centre where the operator has the AI tool open and can use it to offer guidance."

“For our last field visit, we heard the story of a case where HEWs were unable to figure out the underlying cause and asked the AI. The AI suggested typhoid and the family was referred to the nearest health centre. The lab test for typhoid was positive. Cases like these have built a huge amount of goodwill and trust for the solution.”Ravinutala and his team at IDinsight is seeking improvements to the service all the time, including plans to introduce ‘ambient AI’ this year, whereby the AI is listening directly to the HEW on the call, and is “providing suggestions, follow-up questions, and diagnosis on the side”.But with such a powerful tool, there is a constant awareness of the balance that needs to be struck between humans and AI.&nbsp;“One of the core principles is not to replace HEWs or CHWs but to reduce their admin burden and augment their knowledge to provide feedback,” adds Ravinutala. “This principle plays out in numerous design decisions, from what problems we work on to how AI is integrated.”In this vein, IDinsight is showing that where technology respects professional expertise, trust grows. The AI becomes a scaffold for human intelligence, making sure efficiency never comes at the cost of empathy or safety.

A trust-first playbook for intelligent helpThe thread connecting the bee fields of India, the markets of Colombia, and the clinics of Ethiopia is clear. Trust is built through consistent choices over time and it relies on three key humanistic principles.First, AI solutions must keep people in the loop, using technology to augment human decision-making rather than replace it, as seen with IDinsight’s health workers.Second, they prioritize transparent data use, showing people exactly how their information serves them, which lies at the heart of Quipu’s model.And finally, they insist on participation by design, co-creating solutions with the users themselves, just as Buzzworthy did with local beekeepers when creating BeeKind.As these companies continue to grow, they’re proving that the most intelligent help is one people can understand, influence, and trust.

The AI2AI Challenge has demonstrated that the most powerful AI solutions rely on transparency, participation, and putting people first.

Caribou Digital is a research and advisory firm that helps clients change the world by building inclusive and ethical digital economies.

Caribou Digital

Caribou Digital Caribou Digital is a research and advisory firm that helps clients change the world by building inclusive and ethical digital economies.