Loading...
Loading...
Georgia government technology is anchored by three distinct but intersecting ecosystems: the Georgia Technology Authority's Center of Excellence (GTA COE), which serves as the state's AI governance and shared-services hub; the Fulton and DeKalb county governments, which together govern 1.5 million metro Atlanta residents and operate some of the most complex and politically contested government technology programs in the Southeast; and Fort Eisenhower in Augusta, the home of Army Cyber Command, which has made Georgia's Augusta-Aiken region one of the most concentrated cyber and AI talent markets outside the Beltway. Atlanta's smart city program, managed through the Atlanta Mayor's Office of Innovation and Technology, adds a fourth dimension: the city has been among the most aggressive municipal adopters of AI for traffic management, public safety analytics, and citizen services. The interplay between these four ecosystems — state, county, defense, and municipal — creates both the richest government AI talent market in the Southeast and some of its most complex procurement and governance dynamics. LocalAISource connects Georgia government clients with AI specialists who understand GTA's procurement frameworks, the political sensitivities of Fulton County's election and social services AI programs, Fort Eisenhower's cyber defense spillover, and Atlanta's smart city platform architecture.
Updated June 2026
The Georgia Technology Authority Center of Excellence, established in 2021, serves as the state's primary AI governance and capability-building function. The COE operates three programs: an AI use case evaluation service for agencies that want to assess AI feasibility before committing to procurement; a shared AI infrastructure catalog that provides agencies access to AWS and Azure AI services through the state's enterprise cloud agreements; and an AI governance framework that classifies use cases by risk level and establishes documentation and review requirements. Georgia's approach is notably less prescriptive than California's or Colorado's — the GTA COE provides guidance and shared infrastructure but does not require pre-deployment approval for most AI use cases, instead focusing governance energy on high-risk applications (law enforcement, benefits eligibility, criminal justice). The COE's most visible operational success has been the Georgia Department of Revenue's ML-based tax compliance program. GDR administers $30B+ in annual tax revenue and has operated a predictive audit selection model since 2018 — one of the earlier state-level deployments in the Southeast. The model has been retrained annually on audit outcomes and now achieves a 71% hit rate on selected returns, versus the 44% rate the prior statistical sampling approach produced. GDR's data analytics team maintains the model internally, with GTA COE providing cloud infrastructure and ML framework licensing. The Georgia Department of Human Services has been working with the COE on an AI-assisted SNAP renewal processing pilot since 2023. Georgia administers SNAP benefits for 1.4 million residents, and the post-COVID unwinding of continuous enrollment created a backlog that DHS processed primarily through temporary staff augmentation. The AI pilot, covering the 10 highest-volume counties, uses NLP to classify renewal documents and pre-populate case screens — an application modeled on implementations in Florida and Tennessee that GTA's COE benchmarked before recommending the approach. DHS reports early results of 25% reduction in document-processing staff time per renewal in the pilot counties.
Fulton County government administers services for 1.1 million residents — larger than 12 U.S. states — and its government technology programs operate under more intense public scrutiny than most state agencies. Fulton's elections infrastructure, following the controversy of the 2020 presidential election, has been the subject of state legislative oversight, federal monitoring, and independent security assessments. The Fulton County Board of Registration and Elections has evaluated and explicitly not deployed AI-assisted ballot scanning or adjudication tools, instead using AI only for voter registration address standardization and duplicate detection — applications where the audit trail is clean and the legal risk is lower. This is a deliberate governance choice that other Georgia counties have largely followed. Fulton County's courts and correctional system are among the more active users of government AI in the state. The Fulton County Superior Court, which handles a docket of roughly 80,000 annual case events, deployed an NLP-based case routing system in 2023 that classifies incoming filings and assigns them to the appropriate court division — reducing the misrouting rate from 12% to 3% and eliminating a significant source of procedural delay. The Fulton County Sheriff's Office has deployed an ML-based jail population management tool that models detainee flight risk and failure-to-appear probability to support pre-trial release decisions — a system that went through an independent equity audit by the Georgia State University Andrew Young School of Policy Studies before deployment. DeKalb County, with 750,000 residents anchoring Atlanta's eastern suburbs, has focused its government AI investment on property assessment and public works. DeKalb County's Tax Assessor Office deployed an automated valuation model (AVM) based on ML in 2022 — the first Georgia county assessor to use ML in mass appraisal. The model, trained on 15 years of sales data and property characteristics, produced assessment ratios that the Georgia Department of Revenue audit found were more accurate and more equitable across property value deciles than the prior traditional mass appraisal model. DeKalb has since expanded the ML assessment to commercial property, a more complex problem that the assessor is handling in partnership with the Andrew Young School.
Fort Eisenhower in Augusta is the home of Army Cyber Command and the National Security Agency's Georgia operations center — making the Augusta-Aiken corridor one of the most concentrated cyber and AI talent markets in the country. The economic development organization Augusta Economic Development Authority has documented 14,000+ cyber-related jobs in the region, anchored by Army Cyber Command and supported by defense contractors including SAIC, Leidos, Booz Allen Hamilton, and Parsons Corporation. The talent pipeline feeds directly into Georgia state government and Atlanta's city technology programs: the GTA COE's current director and several key technical staff have Fort Eisenhower backgrounds, and Augusta University's School of Computer and Cyber Sciences has a formal research partnership with GTA. Atlanta's smart city program, operating under the Mayor's Office of Innovation and Technology, has been one of the more ambitious municipal AI deployments in the South. The city's PACE (Predictive Analytics for City Engagement) initiative, launched in 2022 and expanded in 2024, uses ML-based citizen service routing, pothole detection from computer vision on city vehicle dash cameras, and predictive infrastructure maintenance modeling on the city's 3,500+ lane miles of streets. The city's Intelligent Traffic Signal System, operated in partnership with the Atlanta Regional Commission, uses AI to optimize signal timing across 175 intersections in real time — the system has reduced average intersection delay by 23% during peak hours on monitored corridors. The Atlanta Regional Commission, which serves as the metropolitan planning organization for 11-county metro Atlanta, has developed ML-based regional growth modeling tools that feed transportation investment decisions, land use plans, and housing policy recommendations for the region's 3.5 million residents. The ARC's AI tools are shared across member counties — giving Fulton, DeKalb, Gwinnett, and Cobb counties access to regional modeling capability that none could afford to develop independently. For vendors, the ARC's shared-model approach is the most accessible entry point into the metro Atlanta government AI market: one sales and deployment effort covers 11 counties and the state DOT simultaneously.
Strategic planning for AI adoption, readiness assessment, and roadmap development
Workflow automation using AI, including Make.com-style automation and RPA
Predictive models, data analysis, and ML pipeline development
Text analysis, document automation, sentiment analysis, and language processing
GTA's COE provides AI feasibility assessments, shared infrastructure access, and a risk-tiered governance framework — but unlike California or Colorado, Georgia does not require COE pre-approval for most AI deployments. Agencies must document AI use cases in GTA's statewide IT inventory and follow the risk classification framework for high-risk applications (criminal justice, benefits eligibility, law enforcement). The COE's practical role is more advisory than regulatory: agencies that use the COE's shared infrastructure and governance templates get faster procurement review and better integration support, but agencies with their own cloud capabilities can bypass the COE process for low and moderate-risk tools. GTA's enterprise cloud agreements (AWS and Azure) provide the fastest procurement path — services within those platforms require only an agency IT certification rather than full competitive procurement.
Fulton County's NLP case routing system for Superior Court went through a legal review by the county attorney's office and a technical review by GTA before deployment — classified as moderate-risk because it routes cases but does not make judicial decisions. The pre-trial risk assessment tool deployed by the Fulton County Sheriff's Office required a full equity audit by the Georgia State University Andrew Young School before the Board of Commissioners approved deployment. The equity audit tested the model for racial and socioeconomic bias using 5 years of historical detention data and found statistically acceptable disparate impact levels — within the thresholds established by the Arnold Foundation's framework, which Georgia uses as its reference standard for pretrial AI tools.
Fort Eisenhower's Army Cyber Command generates demand for AI-capable cyber talent that Georgia-based defense contractors (SAIC, Booz Allen, Leidos, Parsons) service — and those same firms have built civilian government AI practices in Atlanta and Augusta. Augusta University's School of Computer and Cyber Sciences produces graduates who split between defense contracting and civilian government technology roles. GTA has a formal talent pipeline arrangement with Augusta University that provides student internships and faculty consulting access. The most direct spillover is in cybersecurity AI: the AI-powered threat detection tools deployed in GTA's Security Operations Center were developed with input from Fort Eisenhower's defense contractor ecosystem and adapted for civilian state agency environments.
Atlanta's PACE initiative has deployed pothole detection AI (computer vision on city vehicle dashcams, reducing average pothole-to-repair cycle from 22 days to 9 days), an intelligent traffic signal system across 175 intersections (23% peak-hour delay reduction), and ML-based citizen service routing in the city's 311 system (18% reduction in misrouted service requests). The city's smart city platform runs on AWS GovCloud through a Georgia statewide contract. Vendors who want to deploy on the PACE platform need to meet Atlanta's Smart City Vendor Requirements (published by the Mayor's Office of Innovation and Technology) including data minimization standards, API integration specifications, and equity testing documentation. The certification process takes 45–90 days and positions approved vendors for direct procurement from Atlanta and other ARC member counties through the regional shared procurement framework.
DeKalb's automated valuation model uses gradient boosting on 15 years of residential sales data, incorporating 40+ property characteristics plus neighborhood-level economic and demographic features. The GDR audit compared assessment-to-sale ratios across property value quintiles — the standard test for assessment equity — and found DeKalb's ML model achieved a coefficient of dispersion (COD) of 11.2 versus the prior model's 14.8, where the International Association of Assessing Officers recommends COD below 15 for residential assessments. More significantly, the equity test found the ML model had lower vertical inequity (the tendency to over-assess lower-value properties relative to higher-value ones) than the prior traditional model — a finding that addresses a consistent criticism of mass appraisal in Georgia's urban counties.
Browse verified professionals across Georgia.