Machine Learning (ML) and other aspects of Artificial Intelligence (AI) are becoming a critical part of government modernization plans. The fear that "machines will replace people" has largely disappeared. In fact, people see the benefit that ML provides for human workers. ML technology allows machines to do what they are best at - fast computation of large data sets - freeing up humans to do what they do best - analyzing and making sense of the data produced.
The new reality is that while machines will not replace people, those that refuse to adopt and adapt to AI-enabled tools may in fact find themselves replaced by other people that do. The proof is in the pudding. ML tools are helping government teams meet critical challenges of unemployment fraud, natural disasters, racial equity, and healthcare.
Stopping Unemployment Fraud
Fraudulent unemployment claims rose dramatically during the COVID-19 pandemic, costing many states hundreds of millions of dollars and shining a new light on the processes used to determine eligibility and issue payments. Even before COVID-19, Utah was using AI to flag suspicious applications, moving them into a manual review process. In fact, Utah's IT director found himself involved in an issue. He was contacted by human resources to ask if he had resigned. They had received an unemployment claim from "him" and it turned out that fraudsters had applied using his name. The AI technology flagged the claim leading to the HR outreach.
The use of AI can also improve the process for legitimate claims. With AI checking for fraud, identity requirements legitimate claimants have to enter could be less stringent. A driver's license could prove to be enough to check identity rather than also requiring more forms of identification. This removes barriers for people looking to access government services.
Only Machine Learning Can Prevent Forest Fires
Researchers pulled 60 DVDs worth of data from cameras, microphones, rainfall, temperature and humidity sensors, as well as weather and air quality stations to train a Machine Learning algorithm to predict what is seemingly unpredictable - the path and behavior of wildfires. The solution is designed to provide on-the-spot detection, monitoring and analysis of a burning area. This will allow scientists and natural resources officials to use the data to get ahead of forest fires.
Predicting the Racial Make-up of Cities
A team at the University of Cincinnati created a Machine Learning algorithm that can predict segregation changes in neighborhoods. Using Census data, they mapped the racial composition of 300 meter block areas in relation to one another. The algorithm was able to interpret changes within each decade and then make predictions for the racial make-up in 2030. This kind of predictive map could help policymakers, city planners and school districts plan for future services such as housing, transit or Spanish-speaking teachers.
Combining ML and MDs
Machine Learning is being used in the medical community to assist with diagnoses, speeding the process and allowing for care to begin earlier for diseases including cancers, diabetic retinopathy, Alzheimer's disease, and heart disease. Precision and prediction in diagnosis helps increase access to care for all patients and especially those in underserved communities. To make a meaningful impact, a recent GAO report found that key challenges for ML diagnostics include demonstrating real-world performance across diverse clinical settings, developing technologies that integrate into clinical workflows, and addressing regulatory gaps like providing clear guidance for the development of adaptive algorithms.
GovEvents and GovWhitePapers have a wide variety of resources on how the government is using Machine Learning and what is next for the technology.
- ATARC Federal Data and AI Summit (November 17, 2022; Washington, DC) - Artificial Intelligence (AI) and Data Analytics experts explain the importance of AI and Data Analytics across the Federal Government. This event will set out to explain the emerging technology that has been used to strategize Federal AI and Data Analytics, along with the future of AI and Data Analytics in the Federal Government.
- Operationalizing Artificial Intelligence (November 29-30, 2022; Springfield, VA) - Deployment of AI solutions at scale remains challenging. The operationalization of Artificial Intelligence is tremendously complex. This event takes a critical look at what is required to implement GEOINT Artificial Intelligence capabilities for widespread use.
- New & Next: The Government Tech Renaissance (December 8, 2022; Washington, DC) - Explore the innovations driving a tech revolution that is changing government IT. Government leaders and industry experts will discuss the latest emerging technologies and trends, engage in innovative demonstrations, and present the best use cases from across the Federal ecosystem.
- 3 Trends Impacting the Future of Artificial Intelligence in Government (white paper) - This trend report touches on three top trends including AI in today's day and age, how AI affects the government, and what's to come in the future of the Federal workforce.
- AI Bias Is Correctable. Human Bias? Not So Much (white paper) - Individual decisions are shaped by our values, beliefs, experiences, inclinations, prejudices, and blind spots. These "biases" can easily leak into information system design. But overall and over time, modern technology will prove a force for more fair and objective societal actions.
- The Impact of Emerging Technology on AI Within the Federal Government (white paper) - This paper is a summary of a recent roundtable discussion where government IT experts shared the various challenges and solutions they have encountered with the emergence of Artificial Intelligence (AI) technology and advanced data analytics (Data). The group also discussed where they see AI and Data heading in the Federal government, and what steps should be taken for these emerging technologies to be fully accepted and adopted.
For more on Machine Learning in government check out GovEvents and GovWhitePapers.
