Will AI take my job? Navigating AI’s impact on public sector jobs

The question of office automation, robotics and now artificial intelligence has raised numerous concerns and fears about job security in state and local government for many years. However, the technology continues to advance, and humans appear to be more vulnerable than ever.

AI has seemingly emerged from nowhere, and today it is mentioned everywhere. While most don’t feel the least bit threatened, a renewed sense of worry appears to be growing, fueled in part by massive job cuts that have recently begun at the federal level and are now spreading to state and local governments. 

Such concerns seem to come in cycles. In late spring 2017, a website was popularized by the tech media called “Will Robots Take My Job?” The initial data used by the site at the beginning was based on a report titled "The Future of Employment: How Susceptible Are Jobs to Computerization?" which was published by Carl Benedikt Frey and Michael Osborne in 2013. 

The study focused on the susceptibility of just over 700 detailed occupations to computerization. By applying sophisticated mathematical formulas, they concluded that approximately 47% of total U.S. employment is at risk. Now, twelve years later, we have AI to contend with. In 2018, President Donald Trump’s first administration stated that 5% of jobs could be automated entirely.

Reviewing recent White House messaging, AI is framed as a supportive tool, not a replacement for federal employees. They advocate for AI governance frameworks, procurement modernization and boosting internal AI capacity. Conversely, internal experimentation and discussions, particularly in areas influenced by the Department of Government Efficiency, suggest a broader application of AI in administrative roles and consideration of staffing reductions. 

Already, alarms are sounding about the federal government's (mostly DOGE’s) seeming overreliance on AI. Faulty reports have been issued, AI searches have yielded misguided and incorrect data, and there appears to be a lack of human oversight in AI outputs in general.

Anthropic CEO Dario Amodei recently issued a warning that created shock waves across the globe when he predicted that within five years, AI could automate up to 50% of all entry-level white-collar jobs. Others predict that the majority of public sector jobs will remain largely untouched; however, AI and enhanced digital automation may reduce the need for and cost of overtime.

History demonstrates that jobs, whether in the private or public sector, are ripe for automation when they are repetitive and routine. The areas of greatest endangerment are:

• Entry-level administrative positions
• Routine data processing roles
• Basic customer service representatives
• Manual inspection and monitoring jobs
• Simple research and analysis tasks
• Human resource employment screening

There may be a need for fewer employees in such functions, and as AI advances, there will be less need to hire new employees. Seasoned employees will have demonstrated their enhanced productivity through the use of AI. While this may be good news for current public sector employees, it is not good news for future workers.

A majority of economists believe that while AI may indeed replace many government workers, new positions requiring more advanced skills will emerge. Here are some areas that have been suggested:

• AI system administrators and monitors
• Human-AI collaboration specialists
• Data governance and ethics officers
• Digital transformation project managers
• Citizen experience designers
• AI audit and compliance specialists
• Enhanced roles requiring human judgment, creativity, and complex problem-solving

As new opportunities emerge, what happens to those who are displaced? The obvious answer lies in training and retraining. However, as governments are motivated by greater economies of scale, will they be willing to invest in training? How will public employee unions respond? One can only speculate. Meanwhile, after a slow start towards AI adoption, there is much evidence of actual AI applications being initiated. The areas that are receiving the most attention are:

• Administrative and clerical functions: Document processing, data entry, scheduling, basic citizen inquiries
• Regulatory compliance and monitoring: Automated inspection systems, permit processing, tax assessment
• Data analysis and reporting: Budget analysis, performance metrics, policy research
• Customer service: Chatbots for citizen services, automated phone systems, and online form processing
• Predictive analytics: Risk assessment, resource allocation, fraud detection, healthcare patterns

One must also recognize another significant shift in public sector employment: the rise and increasing significance of the vendor community. AI is expensive, and all indicators suggest that costs will continue to rise over time. The overwhelming majority of local governments are relatively small and are struggling to keep up with legacy equipment, maintain and attract qualified IT staff, and as a result, are increasingly turning to outside expertise. Service providers can amortize the cost of AI over multiple accounts, making it a cost-effective alternative. 

One thing is certain: there is no turning back the clock on AI as it is rapidly growing. There are also several potential risks, including the widening of the digital divide and increased economic and skill-based inequalities. There are also ethical considerations, transparency, accountability issues, and public acceptance of robotic and AI technologies.

The next five years will be pivotal as governments adapt, necessitating proactive leadership, investment in people and a commitment to the responsible deployment of AI. Over the next five years, AI is expected to reshape government employment by automating routine tasks, transforming the skills landscape, and creating opportunities for more impactful public service. 

The scale and speed of this transformation require strong leadership, investment in workforce development and a commitment to the ethical and inclusive deployment of AI. Governments that embrace these changes will be better positioned to deliver efficient, innovative, and citizen-focused services. AI will reshape rather than eliminate government jobs, requiring strategic workforce planning. 

Looking back into history, we know that society has withstood enormous economic upheavals due to automation. What is different today is the speed of change and advancement. We have never witnessed such rapid change, which means we must recognize the need for greater and enlightened leadership more than ever. While humans struggle, machines are watching, and unlike any other technology, they are becoming increasingly human-like in their thinking and actions.

Alan R. Shark is an associate professor at the Schar School for Policy and Government, George Mason University, where he also serves as a faculty member at the Center for Human AI Innovation in Society (CHAIS). Shark is the former Executive Director of the Public Technology Institute (PTI). He is a National Academy of Public Administration Fellow and Founder and Co-Chair of the Standing Panel on Technology Leadership. Shark is the host of the podcast Sharkbytes.net.