Working together to apply AI for social good

Millions of people in the Middle East and Africa depend on agriculture for their livelihood. Irrigation is critical to agricultural production in these arid regions, but groundwater resources are being depleted due to agricultural expansion and inefficient irrigation practices. Applying ML to weather and remotely-sensed agricultural data, this team will save water and enhance food production by providing farmers with irrigation schedules that include the precise amount of water needed to irrigate crops in their fields.

Poor environmental practices from illegal mines pollute surrounding areas, contaminating water and fish, and endangering local workers and communities. In Colombia, researchers will use satellite imagery and machine learning to detect illegal mines throughout the country, enabling communities and the government to take actions that will protect people and natural resources.

With over 100 million messages exchanged to date between people in crisis and Crisis Text Line's counselors, it can be challenging to balance spikes in volume and counselor availability. Crisis Text Line will use natural language processing and data on counselor capacity to optimize how they allocate texters to counselors, with the goal of reducing wait times while still ensuring effective communication and deescalation.

Air pollution is a major contributor to poor health and mortality in developing countries. Tracking spatial and temporal pollution patterns is essential to combating it, but can be difficult in low-resource environments. Researchers from Makerere University will apply AI to data from low-cost air sensors installed on motorcycle taxis and other locations around Kampala to help improve air quality monitoring and forecasting and inform future interventions.

Ambulances are often the first point of contact with someone who is suicidal, making ambulance clinical records a unique data source to help inform suicide prevention efforts. By using AI tools to analyze these records, Turning Point, a national center within Eastern Health, will uncover critical suicide trends and potential points of intervention to better inform policy and public health responses.

Low-income students frequently do not receive the feedback and practice needed to develop strong writing skills, and, as a result, 27 million low-income students in the United States are not proficient writers. Quill.org will use structured deep learning to provide students with immediate feedback on their writing, enabling students to revise their work and quickly improve their skills.

Enabling medical staff to prescribe the right antibiotics for bacterial infections is critical for individuals and increasingly for global management of antibiotic resistance. Antibiograms help guide these decisions, but in low-resource areas, medical staff often lack the skills to interpret them. Using image recognition and a smartphone app, the Fondation Médecins Sans Frontières will help staff analyze anti-microbial images and provide appropriate patient treatment.

As concerns about mis- and disinformation grow, the work of fact-checkers has never been more important. Together with leading fact-checking organizations AfricaCheck and Chequeado, Full Fact will use AI to provide trend monitoring and clustering tools to aid fact checkers’ analysis of news and other information. This will give fact-checkers more time to focus on research, analysis, and writing articles that contextualize the news and help all of us make more informed decisions.

Millions of tons of plastic are dumped into our oceans each year, much of it coming from communities where waste management is under-resourced or absent. Gringgo is creating an image recognition tool to help informal-sector waste collectors and independent waste management companies increase recycling rates and better integrate with city sanitation crews. This will improve and expand community trash collection and reduce ocean plastic pollution.

Many deaf and hard of hearing individuals in Brazil struggle to read and write, and they rely largely on sign language to communicate. Hand Talk uses AI to translate Portuguese into Brazilian Sign Language using a digital avatar named Hugo. This tool enables digital communication for deaf and hard of hearing Brazilians, and also increases Brazilian Sign Language learning.

Human rights lawyers are currently required to sift through vast document repositories to identify the most relevant facts for their case. HURIDOCS is using natural language processing and machine learning methods to extract, explore and connect relevant information in laws, jurisprudence, victim testimonies, and resolutions. With accessible global documents, human rights lawyers can more efficiently and effectively research and defend their cases.

In an emergency, a matter of seconds can drastically influence outcomes in saving lives. In this project, researchers at Columbia University will partner with the New York City Fire Department analytics team to build models that will optimize first responder responses to the 1.7 million emergencies the department answers each year, accounting for factors such as weather, traffic, location, type of emergency, and more.

The storage condition of a vaccine can significantly affect its effectiveness, which is especially challenging in remote regions with limited infrastructure. Nexleaf tracks data on vaccine cold chain systems, providing critical information to health workers and other decision-makers on the system’s functioning. They will use AI technologies to build data models that predict vaccine degradation, quantify the value of vaccines at risk, and ultimately develop an end-to-end system to ensure safe, effective vaccine delivery.

Rainforests are under increasing threat from illegal logging and global warming. Rainforest Connection is using commonplace mobile technology and deep learning for bioacoustic monitoring to detect immediate threats and track rainforest health. The network roots out illegal logging activity and other environmental dangers while simultaneously analyzing the health of the rainforest, making the data available for any scientist around the world.

Helping refugees find jobs is a key priority for European cities, but translating refugees' skill sets to the European labor market is not always straightforward. Skilllab helps municipal employment services integrate refugees into local job placements. Using an app powered by AI, refugees can document their skills quickly and in their native languages, and receive recommendations of relevant career pathways to explore.

Parental involvement at home and at school is critical to students' educational outcomes, but building relationships with teachers can be challenging for families facing economic, educational, cultural or language barriers. TalkingPoints will help break down these barriers by using AI to enable two-way translated communication and personalized coaching content to guide parents' engagement with the teachers and at home with their children.

Every year, landslides cause thousands of preventable deaths around the world, and can displace entire communities and destroy habitable lands. Collaborating with the U.S. Geological Survey Landslide Hazards Program, Penn State researchers will use deep learning tools to better predict landslide location, timing, and impacted areas and create a warning system to minimize the impact of these natural disasters.

In the US alone, an estimated 1.5 million LGBTQ youth seriously consider suicide or experience a significant crisis every year. The Trevor Project saves lives by supporting at-risk LGBTQ youth via phone, text, and chat. Using natural language processing and sentiment analysis, counselors will be able to determine a LGBTQ youth’s suicide risk level, and better tailor services for individuals seeking help.

More than a billion people live in smallholder farmer households worldwide, and many of these farmers struggle with avoidable pest damage that can wipe out up to 50%% of annual crop yield. In this project, as farmers and agriculture program workers take regular photos of pest traps, AI models on their phones classify and count the pests. The data will be used to provide millions of farmers with timely, localized advice to reduce pesticide usage and improve yield.

Billions of dollars have been spent building monitoring systems to measure the emissions of major fossil fuel power plants. These monitoring systems make critical emissions reduction initiatives possible but are not accessible in communities that cannot afford them. WattTime will use image processing algorithms and satellite networks to replace expensive, on-site power plant emissions monitors with a globally accessible, open-source monitoring platform.