Boon

Refugees living in camps lack access to stable work and therefore rely on an increasingly small pool of aid money to survive. The few work opportunities that do exist are highly competitive and often don’t pay living wages. Boon is developing a mobile app that will connect refugees through their smartphones to digital work opportunities. This will turn every smartphone into a revenue-generating system, allowing refugees to do dignified work specifically adapted for their skills. By reimagining refugees as people willing to support themselves rather than merely aid recipients, we can help empower the 25 million refugees worldwide to lift an entire generation of displaced people out of poverty. 

There are 25 million refugees worldwide, and 15 million of them are in protracted situations of more than 5 years. All displaced people were guaranteed a right to work by the 1951 UN Convention, but most still lack access to employment opportunities - especially ones that offer training or upward mobility. At the same time, UNHCR has struggled to keep pace with the growing number of displaced people. Their mandate has doubled since 2012, but as of 2018 they faced a budget deficit of $4.5 billion, almost 45% of their annual budget. These budget cuts have forced refugees to live with fewer resources and basic necessities, while host countries and international communities are strained to fundraise continuously to provide the needed aid. With few traditional employment opportunities available, and despite high smartphone penetration in many camps, refugees are still unable to connect with the growing global demand for remote work. Lack of employment reduces health and quality of life for refugees and increases the reliance on an outdated system of international aid.

Boon is a mobile marketplace that connects refugees to digital work opportunities. In 2019 there was $92 billion dollars of outsourced work and the demand is only growing. Boon’s team will secure contracts for digital work and break it into discrete work packages for deployment on the app. Refugees will be able to download these work packages and complete them offline according to their schedule. Similar to Amazon’s MTurk platform, the users will be paid on a per-task basis that can cumulatively bring in a steady income, tapping into local mobile money platforms like Mpesa for payment distribution. In addition, as users progress through simpler tasks like dataset creation for AI or ML programs, they will be encouraged to take on more complex and higher-paid tasks like content moderation or transcription. This work will not only provide life-changing income and stability, but will provide them skills necessary for other work opportunities in an increasingly digital world.

Companies: Most large corporations already outsource work, but are looking for opportunities to set them apart in a world where consumers increasingly make purchasing decisions based on a company's social impact. 

Refugees: Providing refugees with a dignified, stable job will give them financial independence and increase their quality of life. In addition, working with Boon will give them the skills they need to compete for other jobs in the digital workforce.

Boon provides skills training as part of the onboarding process for every refugee user, through an interactive, hands-on program. They will learn how to conduct a series of digital tasks, such as data identification for AI programs, which are applicable to many digital careers. They will also unlock more access to training as they progress through the work. We will reinforce this by explaining to users how their tasks relate to the bigger picture, which will both increasing refugee engagement in the process and educate them about modern digital work.

Micro work is an established industry in many parts of the world. However, traditional models typically rely on full-time employees working on a computer at a centralized location. This model has high labor and overhead costs, and their work is intrinsically limited to the number of computers they can put into a room. Boon’s innovation is to break away from the desktop model and launch remote work into the digital crowdsourcing future. By harnessing the power of smartphones, we will leverage the enormous refugee workforce to process more work faster. In addition, diverse workforce will focus on ethical AI, an increasingly crucial aspect of dataset creation. Samasource, based in Kenya, is a prime example. Their model pioneered early socially-based remote work, and they have carved a niche for themselves by having full-time staff in locations based around the world. However, even they are limited to computer centers, which have a fixed number of operators. Boon solves this model by crowdsourcing the microwork to thousands of refugees, allowing us to handle larger datasets at a more reasonable cost.

Boon is a technology-driven marketplace, accessible via mobile application for smartphones. The application is a hub that connects a customer-facing platform for task uploading, with a user-facing app for task completion. The hub will use AI to evaluate users and assign them appropriate tasks based on their ability and availability. In addition, the tasks themselves will be used to power AI and ML programs around the world through creation of precise datasets. Similar to Scale AI, which has developed AI engines for imagery, computer vision, and sensor data, aided by 30,000 contractors to refine the dataset, Boon will develop robust AI/ML engines to process the initial datasets with trained refugee workforce to validate and refine the program accuracy. Boon's advantage over companies like Scale AI is the global footprint in resources and a philosophical emphasis on ethical AI (e.g., combating racial bias), an area that has room for improvement even for established AI players, exemplified by the current political climate.

Digital outsourcing is an established and rapidly-growing industry, responsible for $92 billion USD in revenue in 2019 alone. An established example of remote tasking is Amazon’s MTurk platform, which provides a marketplace for everything from data validation to content moderation. Mturk has been operating successfully with this technology since 2005, with tens of thousands of users. Labelbox is another example of a company that manages labeling projects (e.g., allowing multiple labelers to work on the same project at once). The technology that Boon would need to create the platform, develop the AI/ML algorithms, and create the supporting program tools exists across several players. Boon has the opportunity to be the one-stop solution for customers, with added benefits of supporting a social enterprise.

Benjamin and Catherine, two of Boon's co-founders, worked in Kenya for five years providing sanitation services for local towns, including a refugee settlement with 180,000 people. While trying to address sanitation issues, they discovered that the root of many of the refugees' issues is poverty, with less than 10% of adults employed due to a lack of opportunities. The people living in refugee camps deserve reliable employment in order to provide for themselves and their families. Providing remote microworking opportunities empowers them to be economically self sufficient while training them in digital literacy, preparing them for the modern workforce. A recent impact study noted that workers’ income increases between 40 and 200% as a result of microwork. We need to rethink the opportunities available to refugees around the world by empowering them to break the cycle of poverty and aid-reliance. 

The AI industry is growing rapidly, just like our dependence on it as a society, and our customers require humans to validate the data so they can train their algorithms. In an increasingly socially aware global context, these companies are searching for ways to fulfill their growing data demands while ensuring the data is ethically refined. Providing a service that guarantees efficiency and ethically-worked data will help these companies build the algorithms they need to power their work.

Boon is not currently serving any people, but we plan to have a direct impact on 50,000 people within 1 year (10,000 people at 5 pilots each), and more than 1,000,000 people within 5 years (50,000 people at 20 camps).

Based on our ties with academic institutions such as Stanford, Northwestern, and Emory, we aim to pilot 5 projects with academic research centers and small-scale tech companies in the first year. We will build on the success of initial contracts to grow our reputation and build lasting relationships in the private sector that will unlock more opportunities for growth. Within five years we will be a household name in microwork and have a strong, dedicated workforce distributed in refugee camps throughout the world. We will galvanize the humanitarian industry by forcing the major players to rethink their approach to livelihoods and employment for refugee populations. The taskers and their families will have driven millions of dollars to flow through the camps, boosting local economies and bringing stability to a typically untenable situation. For AI companies, we hope to expand beyond providing labeling services to actively leveraging our global workforce to define and collect datasets in five years. Even as our AI/ML algorithms improve and require less human oversight, we will actively create structured data in anticipated areas of need to ensure a steady flow of work for refugees.

Boon faces a few key barriers. 1) Geography, context, and language. Using distributed refugee populations will mean that we will have to have the training available in multiple languages, and potentially adapt tasks to make sense in different cultural contexts. 2) Building a coordinated and asynchronous platform. The work of parcelling out tasks and distributing them will be an interesting challenge to overcome technically, and at the center of Boon’s model. Answering the question, “How can we intake datasets with millions of data points and distribute them among different applications.” 3) Payment. Working with international money transfers and ensuring users are connected to mobile banking will be a unique hurdle for Boon’s work. 4) Digital literacy and soft skills. While microwork generally involves simple and repetitive tasks, digital literacy may not be widespread. Additionally, there may be varying levels of soft skills (e.g., communications, creative problem solving) among taskers.

1) In order to adapt our platform to multiple languages, we can use our own labor force to translate the training programs, drawing on the bilingual abilities of many refugees. We can turn this challenge into a benefit by understanding their cultures and drawing on their differences to enhance our product offerings and not limit ourselves to English-only datasets. We will introduce new languages in cohorts of refugee populations (e.g., English first, Swahili next) to ensure smooth transition. 2) Building a coordinated and asynchronous platform. We will be pursuing technical partnerships to help us build out the back-end capacity to allow for parcelling tasks and distributing them with low data usage. 3) Payments. We will be seeking to build partnerships with organizations experienced in mobile banking among displaced populations to draw on their knowledge of financial transactions to smartphones. 4) Digital literacy and soft skills. Part of Boon’s mission is to help elevate digital literacy and increase livelihood for refugees. We will partner with virtual learning platforms (e.g., Lynda, Coursera, digital literacy NGOs) to provide robust training for workers.

NA

Full-time: 2

Part-time: 4

Our team has deep and diverse experience across a range of relevant industries, from humanitarian to technology. Each person has worked in a startup environment and understands what it takes to succeed. 

Benjamin Kramer. Helped grow a venture-backed startup from 2 to 90 people in Kenya, with 6 years of experience working in developing countries and refugee camps. He will guide the strategic growth. Currently a pre-MBA at Stanford GSB.

Vickie Chi. Vickie has led startup marketing and product, most recently at GRAIL, a $2B AI-driven biotech company. Previously a consultant, she launched an AI transformation lab with Google for Fortune 500 executives. Currently a pre-MBA at Stanford GSB; she will drive growth and partnerships. 

Diego Hakspiel. Diego has worked in the private and non-profit sector, leading innovation in refugee settings in several countries. He will guide the humanitarian aspect of the work, drawing on his deep contacts within the industry. 

Anderson Riungu. A full-stack developer with 13 years experience as an engineer, including launching technology companies using AI. He started SimplePOS, a point-of-sale app with >10,000 users. He will lead the technology aspects. 

Catherine Berner. An environmental engineer who has led large technical projects in Kenya for 5 years. Currently a pre-MBA at Stanford GSB, she will run our global expansion.

Dillon Barr. Dillon is an experienced sales professional with global experience in developing countries, including selling pay-as-you-go solar home systems. Doubling as a trained sales and professional coach, he will mentor our sales and leadership team.

N/A

The Boon app will meet the demands for two groups: companies and refugees. Companies require large volumes of data to be generated or scrubbed, in a professional and timely manner. Common contracts are for image tagging, AI data scrubbing, content moderation, and transcription services. Work of this volume requires access to both a large labor pool and a platform that can ensure the highest quality outputs. Boon fills this need by tapping into the enormous potential for refugee workers and by leveraging powerful algorithms to correlate and validate their work. In addition, we will develop our own AI/ML models with our workforce to supplement and expand our capacities. At a time when many companies are seeking to build a socially powerful brand, they will be looking for opportunities to partner with organizations driving social impact without sacrificing on quality or price. At the same time, millions of refugees are eager and able to work, but traditional jobs are limited. We fill both needs by connecting them in a central hub and dispersing the work according to ability. Similar to other digital marketplaces, we will take a percentage of each contract as a brokerage fee. By having a crowdsourced labor pool of contractors, we will keep labor and overhead costs low.

To launch the application and run the pilot, we will first seek to exhaust grants for social enterprises and humanitarian organizations. As a for-profit social company, we will be able to take a blended capital approach for subsequent funding rounds, tapping into equity and debt as well as large humanitarian contracts. Our revenue will come from taking a percentage of each contract as a brokerage fee, and will soon cover our expenses. Additional revenue sources will be available at scale, including getting contracts for services from UNHCR or INGOs, and connecting talented taskers with companies for remote employment.

MIT’s Solve is designed by and for builders like our team at Boon, and we see incredible opportunity in being part of the community of Solvers. Our team is ambitious, and the mentorship offered for both technical aspects and growth strategy, will unlock our potential to create impact through microwork. We will draw on Solve’s world-renowned technical experts to develop a platform that can thrive even in cross-border, remote environments. In addition, our work is strongly linked to partnerships with cutting-edge technical companies on the forefront of AI, and Solve’s network will be an enormous boost to our team’s contacts. A typical challenge for new companies is getting validation, and having Solve’s media attention and stamp of approval will open doors across the country

While we will employ existing technology, our unique combination of the technology combined with the particularly nuanced market of refugee camps will require high-level technical creativity. We would benefit from having mentorship to grow and develop our technical team to meet these high demands. 

We are seeking technical partners to help guide and develop the systems which drive our platform, specifically those with experience in cross-border financial transfers, AI algorithm development, and online marketplaces. Furthermore, we are actively seeking to partner with organizations who want to prioritize ethical dataset creation without sacrificing quality. For individuals, we would greatly benefit from mentorship from digital product managers, back-end system developers, and people working closely with developing AI/ML algorithms. For our educational components, we are looking for partnerships with Lynda, Coursera, or other digital literacy platforms. 

At Boon we believe that the key to self-reliance is financial independence. That is why we are creating a platform designed specifically for refugees to find reliable and well-paying work. Armed with a smartphone, any refugee will now be able to advance their career and earn a stable revenue while simultaneously being trained to compete in the global digital workforce. We understand the unique demands on refugee life in a camp, and are including that perspective in all of our designs. 

Economic independence is one of the most effective ways to empower women and girls. By providing an opportunity for women to work by themselves using only their smartphones they will gain control of their own finances, which is especially crucial in places like refugee camps where employment is already limited. They will be up-skilled through our training programs and will be able to leverage those skills if they want to gain other employment in their future as part of the global digital workforce. 

Our platform will not only provide employment leading to economic freedom, but through our educational trainings and the nature of the work itself, it will give refugees the chance to gain crucial hands-on experience in the modern digital workforce. Our program also ensures taskers understand how their work impacts the algorithms they support, which is a unique opportunity for many refugees to be included in high-demand AI workplace. As the refugees build skills and experience, they are given more complicated tasks, leading to steady professional development over time.  

By creating a platform to connect refugees to microwork opportunities, our team is at the very nexus of refugees and AI. We will be bringing employment to tens of thousands of people in refugee camps and including them in the AI/ML dataset marketplace by channeling some of the benefits of this incredible technological advancement to those who need it most. Our passionate and dedicated team will use the AI for humanity Prize to onboard talent and launch our pilot programs in camps around the world. The prize will go directly towards building the systems needed to empower refugees through AI.