Welcome to the awesome list of uses of AI in government! Below are projects that contributors have read about, participated in, or believe are feasible based on experience with related applications. If you would like to add or revise an item on the list, please send a Pull Request through GitHub.
Potential Agency | Application | Technology | ||
---|---|---|---|---|
Agriculture | Macro crop / tree / plant identification | Supervised ML and computer vision | ||
Agriculture | Micro leaf / fruit / other feature identification for harvest, pruning, etc. | Supervised ML and computer vision | ||
Agriculture | Weed / pest identification | Supervised ML and computer vision | ||
Agriculture | Erosion / sand / rock identification | Supervised ML and computer vision | ||
Agriculture | Animal - breed (wanted or unwanted) / disease / other health detection | Supervised ML and computer vision | ||
Agriculture | Robotic farming (combines several of the above) | Supervised ML and computer vision | ||
Agriculture | Fertilizer / pesticide recommendation and optimization | Supervised ML | ||
Agriculture | Precision irrigation | Supervised ML and computer vision | ||
Agriculture | Meteorology and effects on crop yield | Supervised ML | ||
Agriculture | Market forecasting | Supervised ML | ||
Agriculture | Crop / animal yield forecasting | Supervised ML | ||
Agriculture | Food grading (fruit, meat, etc.) | Supervised ML and computer vision | ||
Agriculture | Food label inspection | Supervised ML and computer vision | ||
Commerce | Prediction of individual survey response or survey response rate | Supervised ML | ||
Commerce | Economic / business forecasting | Supervised ML | ||
Commerce | Fuzzy matching of trademarks / business entities | Supervised ML | ||
Commerce | Classification / triage of industry and speciality for patent applications | Natural language processing and supervised ML | ||
Commerce | Prediction of trademark/patent application status (e.g. approved, clarification) | Natural language processing and supervised ML | ||
Commerce | Commercial fishery - detection of fish / boats (including unexpected boats) | Supervised ML and computer vision | ||
Commerce | NTIS - offers AI services to any agency | Many | ||
Commerce | Census response | Supervised ML | ||
Commerce | Census anomaly detection | Unsupervised ML | ||
Commerce | Meteorology and weather predictions (NOAA, NWS) | Supervised ML | ||
Commerce | Determine species composition, length and count of fish from commercial fishing | Supervised ML and computer vision | ||
Defense | Signal/image classification - identify vehicles, buildings, weapons, and other objects from ground, aerial, or underwater images and sensor readings (e.g. LIDAR) | Supervised ML and computer vision | ||
Defense | Predictive maintenance - proactively suggest maintenance based on patterns seen in equipment | Supervised ML and (optionally) computer vision | ||
Defense | Automated sensor fusion (automatically synthesizing data from multiple sources into a single view) | Many | ||
Defense | Missle / mortar / bullet source identification through visual / radio / audio signal | Supervised ML and computer vision | ||
Education | Predicting school / student success | Supervised ML | ||
Education | Loan repayment probability | Supervised ML | ||
Energy | Natural disaster forecast | Supervised ML | ||
Energy | Energy demand and supply forecasting (improved efficiency, reduced costs and emissions) | Supervised ML, Time series forecasting | ||
Energy | Simulation and other physics research | Many | ||
Energy | Wildfire prediction and containment | Supervised ML | ||
Energy | Prediction of breakdowns and maintenance required | Supervised ML | ||
Environmental Protection Agency | Detection of unreported leaks / pollutants | Supervised ML and computer vision | ||
Environmental Protection Agency | Superfund site identification and remediation recommendations | Supervised ML and computer vision | ||
Equal Employment Opportunity Commission | Pre-adjudiciation of discrimination claims | Supervised ML and natural language processing | ||
FDA | Adverse event anomaly detection (e.g., spikes in consumer complaints about a drug) | Unsupervised ML | ||
FDA | Food grading (eggs, etc.) | Supervised ML and computer vision | ||
FDA | Pill identification and screening for anomalies/debris | Supervised ML and computer vision | ||
FDA | Food/drug label inspection | Supervised ML and computer vision | ||
Federal Deposit Insurance Corporation | Forecast of bank distress/ failure | Supervised ML and natural language processing | ||
General / All | Fraud detection | Supervised and unsupervised ML | ||
General / All | Record linkage / deduplication | Supervised ML | ||
General / All | Reading text from hand-written forms | Computer vision - optical character recognition | ||
General / All | Cybersecurity threat identification (e.g. malware, network probe) | Supervised and unsupervised ML | ||
General / All | Helpdesk automation / chatbot | Natural language processing | ||
General / All | Auto-summarization of regulations for potential reduction / combination | Natural language processing | ||
General / All | Auto-summarization for rationalization of SOPs | Natural language processing | ||
General / All | Extract key information and provide structure for unstructured data | Natural language processing | ||
General / All | De-identification of sensitive data (e.g. for FOIA requests, analysis) | Natural language processing - named entity recognition | ||
General / All | Imputing missing values in data sets | Supervised ML | ||
Health | Predict health outcomes from patient/case data and design interventions to help avoid it proactively | Supervised ML | ||
Health | Predicting over/under utilization of services from case/patient data | Supervised ML | ||
Health | Identification of high/low risk patients for differential treatment (e.g. proactive care) | Supervised ML | ||
Health | Detect conditions from wearable devices (e.g. Parkinson's from gait), diabetes from blood monitor | Supervised ML and unsupervised ML | ||
Health | Prescriptive analytics for health improvement / life coaching | Supervised ML | ||
Health | Diagnostic assistance from structured (e.g. patient history) and/or unstructured (e.g. lab imaging) data | Supervised ML and unsupervised ML | ||
Health | Disease prognosis (e.g. likely to recover or decline) | Supervised ML | ||
Health | Physician fatigue mitigation | Natural language processing of medical charts , voice to text of clinical notes | ||
Health | Streamline diagnosis by analyzing structured (EHR) and unstructured medical data (e.g. radiology images, blood tests, EKGs, genomics) | Natural language processing and supervised ML | ||
Health | Using image recognition to speed up insurance claims processing | Supervised ML, natural language processing, and computer vision | ||
Health | Using AI to profile and explore hospital efficiency | Supervised ML | ||
Health | Patient matching (a form of record linkage) | Supervised ML | ||
Homeland Security | Pre-adjudication of visa/ immigration applications | Supervised ML and natural language processing | ||
Homeland Security | Identification of threats to federal facilities / workers inside the US | Supervised ML and computer vision | ||
Homeland Security | Identification of imports for customs duties and threat identification | Supervised ML and computer vision | ||
Homeland Security | Automated screening of package / freight exterior with imaging | Supervised ML and computer vision | ||
Homeland Security | Automated screening and detection of objects in package / freight interior | Supervised ML and computer vision | ||
Homeland Security | Prediction of disaster response requirements | Supervised ML | ||
Homeland Security | Predictive policing | Supervised ML and unsupervised ML | ||
Housing and Urban Development | Identification of bias in rent/homeownership/mortgage programs | Supervised ML | ||
Interior | Safety issues at oil/gas sites | Supervised ML and/or computer vision | ||
Interior | Invasive species detection | Supervised ML and computer vision | ||
Justice | Predicting recidivism | Supervised ML | ||
Justice | Predicting risky inmates | Supervised ML | ||
Justice | Detection of networks / criminal rings | Supervised ML and unsupervised ML | ||
Justice | Gunshot / firearm detection | Supervised ML | ||
Justice | Case review for faults/oversights | Supervised ML and natural language processing | ||
Labor | Predicting industrial accidents by location | Supervised ML | ||
Labor | Forecasting the severity and impact of accidents / workplace violence/ safety issues | Supervised ML | ||
Labor | Predicting violations by site (e.g. from required inspections) | Supervised ML | ||
Labor | Automatically detecting hazards at work sites (e.g. uncovered containers, leaking dam) | Supervised ML and computer vision | ||
Securities and Exchange Commission | SupTech - automated supervision of regulatory compliance requirements (e.g. reviewing 10Ks) | Supervised ML and natural language processing | ||
State | Program efficacy and program participant response | Supervised ML and natural language processing | ||
State | Translation | Natural language processing | ||
State | Social media monitoring for sentiment | Natural language processing | ||
Transportation | Tax Trade Bureau - Alcohol Labeling inspection | Supervised ML and/or computer vision | ||
Transportation | Detection of birds in the runway | Supervised ML and computer vision | ||
Transportation | Airport runway pavement inspection | Supervised ML and computer vision | ||
Transportation | Railroad track inspection | Supervised ML and computer vision | ||
Transportation | Highway and bridge inspection automation | Supervised ML and computer vision | ||
Transportation | Auto Safety - Lane keep assist systems and similar | Supervised ML and computer vision | ||
Transportation | Oil/gas/chemical pipeline hazard inspection | Supervised ML and computer vision | ||
Transportation | Suspicious personnel on train track right of ways | Supervised ML and computer vision | ||
Treasury | Predict likelihood of succesful/appealed/overturned audit | Supervised ML | ||
Treasury | Predict likelihood of substantive error in return | Supervised ML | ||
Treasury | Predict which contact/response method and message will get a response from a taxpayer | Supervised ML and natural language processing | ||
USPS | Handwriting recognition on hand-addressed mail | Supervised ML and computer vision | ||
USPS | Route optimization | Supervised ML and computer vision | ||
Veteran's Affairs | Job / skill matching based on veteran profile / resume | Supervised ML | ||
Veteran's Affairs | Matching patients with clinical trrials based on medical history | Supervised ML | ||
Veteran's Affairs | (see Health) |