Outside the US, the definition of poverty is income of $1.50 a day or less per individual. In the United States the definition of poverty is income less than $34 per day per individual or a family of four with income less than $69 per day. After much research, experts have agreed that low income is only one indicator of poverty and a more accurate indicator is the measurement of consumption. It is necessary to fully understand what factors affect well-being therefore measuring consumption was determined a more accurate indicator of poverty.\nTo know how to find solutions to poverty, we must first define the parameters and scope of poverty. The causes of poverty, the location of poverty and measuring consumption are factors involved in determining well-being and by extension a few of the data points that govern well-being. Next through the process of measuring on a previously defined and acknowledged scale, a threshold determination is agreed upon.\nAlthough inclusion of many necessary data points is required, the need for practicality and simplicity are necessary for this article. Let\u2019s just say that poverty is caused by the lack of education and life skills and the lack of access to food and clean water which can be regionally systemic, an act of a disastrous occurrence or war or both. Poverty could also be caused by over population.\nSatellite imagery plus economic variables pinpoints poverty\nIn many countries, a full set of economic variables could be extremely unreliable as input data because many 3rd world countries have a weak infrastructure, a lack of thoroughness to gathering the necessary data and a lack of cooperation in sharing the data. Political power structures have their own agenda so the data could have a tendency to show much more optimistic numbers or made-up numbers. Overall no matter the country or location, the lack of quality data hinders development and economic growth for the region.\nThrough streaming satellite imagery, researches are identifying geographical regions of poverty more accurately by identifying dense concentrations of light at night as affluent. Experimentally, Stanford University\u2019s Marshall Burke compiled night-time and day-light images of many areas of Africa (Rwanda, Nigeria, Uganda, Malawi and Tanzania) pinpointing specific key performance indicators for a gap analysis.\nBy including economic data along with the geospatial data accumulated as input into an AI\/ML system, Burke and his team were able to predict regions of poverty within 81%-99% accuracy. This level of accuracy can positively influence economic aid through the administration of goods and services more precisely therefore reducing costs and helping more people. Also, education can be overseen and managed more effectively.\nCan you imagine smart chatbots substituting for teachers in poverty-stricken areas? Mobile trucks with educational classrooms power AI\/ML systems of ongoing structured educational level progression based on reoccurring skills assessments. As long as access to computers and the internet are available which may include satellite communication networks, AI-teachers can provide students education based on a controlled syllabus.\nInequality for good education can be eliminated because of a reduction in labor costs, elimination of the tuition (money) barrier and allowing easy access to proliferating data in a structured format. Tailored learning situations can be individualized and optimized and teaching methods for less knowledgeable backgrounds can have equal access and level the educational playing field.\nTech giant IBM is also looking at different ways to alleviate poverty through the application of AI\/ML, along with investigating other societal issues. Their Science for Social Good directive and the Literacy Coalition of Central Texas (LCCT) partner on a project called Simpler Voice to overcome illiteracy. Visual cues are how low-literate adults and children as well as AI\/ML systems recognize and comprehend information (learn).\nThe mobile app, Simpler Voice integrates IBM Watson natural language using text-to-speech services with novel image generation code. AI\/ML phrasing through generative adversarial networks (GAN) provide alternative conceptions on a smart phone turning text into simple verbal messages. The AI\/ML system can parse complex text of public signs, textbooks, manuals and even short spoken recordings creating either text to speech or speech to text conversion.\nA pilot test was performed at a grocery store on shampoo bottles, canned goods and dishwashing detergent. An LCCT student unable to read the box of dishwashing detergent to recognize the product or the price scans the barcode. Simpler Voice interprets the barcode product description and price along with displaying a picture of someone using the detergent. Simpler Voice interprets key words and phrases to verbalize \u201cdishwasher detergent\u201d and display a picture of a person loading a dishwasher with the detergent. It may also verbalize who should use the detergent, how it should be used, child warnings and safety information.\nAnother test performed on prescription medications verbalized how to take it, safety and refill information along with warnings for possible allergic reactions. Simpler Voice\u2019s roadmap will expand to include legal documents, medical documents and service agreements.\u00a0\nGrowing resilient crops increases food supply\nIn anticipation of the predicted 2050 9.6 billion global population, Carnegie Mellon University (CMU) has focused on eradicating poverty using robotics and AI\/ML within agriculture. CMU\u2019s project FarmView is researching the crop growth expansion of sorghum and other staple food crops in developing 3rd world countries by using AI\/ML along with robots such as drones. Sorghum is used as a food source and in the manufacturing of biofuels. It is the fifth most important protein enriched cereal crop grown in the world and because it has over 42,000 varietals, it is genetically resilient even when planted and grown in less than ideal circumstances.\nDrones capture data that is analyzed for optimum planting and harvesting strategies. The hope is that through AI\/ML sensors a comprehensive plant breeding and crop management system will speed the growth cycle of a \u201cdrought and heat-tolerant grain\u201d planted in \u201cfamine-stricken\u201d areas providing more food for the region therefore increasing their main source of income.\nIBM\u2019s Science for Social Good program along with many other organization\u2019s programs are expanding their uses of AI\/ML to investigate illiteracy and the current and predicted food shortage. The goal is to improve people\u2019s well-being and to eradicate poverty through reducing hunger, expanding food distribution, growing a heartier food supply and making education available to those who until AI\/ML lacked access. A preamble to the United Nations Sustainable Global Agenda explains the following observations and motivations:\n\n\u201cThis Agenda is a plan of action for people, planet and prosperity. It also seeks to strengthen universal peace in larger freedom. We recognize that eradicating poverty in all its forms and dimensions, including extreme poverty, is the greatest global challenge and an indispensable requirement for sustainable development. All countries and all stakeholders, acting in collaborative partnership, will implement this plan. We are resolved to free the human race from the tyranny of poverty and want and to heal and secure our planet. We are determined to take the bold and transformative steps which are urgently needed to shift the world onto a sustainable and resilient path. As we embark on this collective journey, we pledge that no one will be left behind.\u201d\n\nAI and ML can be instrumental in the future of mankind\u2019s healthy well-being and quite possibly survival on the earth.