Sections

Commentary

How federal layoffs set the stage for greater privatization and automation of the US government

January 30, 2025


  • The Trump administration’s recent executive orders aim to reduce the federal workforce, potentially leading to increased privatization and automation of government services.
  • Influenced by Big Tech leaders, these changes raise concerns about data security, bias in AI systems, and the digital divide, which could leave many Americans without access to essential services.
Elon Musk greets U.S. President-elect Donald Trump as he arrives to attend a viewing of the launch of the sixth test flight of the SpaceX Starship rocket, November 19, 2024. Brandon Bell/Pool via REUTERS/File Photo
Elon Musk greets U.S. President-elect Donald Trump as he arrives to attend a viewing of the launch of the sixth test flight of the SpaceX Starship rocket, November 19, 2024. Brandon Bell/Pool via REUTERS/File Photo
Editor's note:

A minor edit was made to this article for clarity on January 31, 2025. 

In his first two weeks of office, President Trump signed several Executive Orders (EOs) to fulfill one of his many campaign promises—to reduce the size of the federal government. He has rolled back diversity, equity, and inclusion (DEI) initiatives, asserting that the federal government will no longer consider race, ethnicity, or other federally protected characteristics in hiring and retention decisions. In recent days, he announced a financial buyout to federal employees who do not wish to comply with the new Return to Office (RTO) mandate, which requires employees to be in an office for five days per week, despite concerns about available office space. The details of the buyout were outlined in an email with the subject line, “Fork in the Road,” sent by the Office of Personnel Management (OPM) on January 28, 2025, to over 2 million federal workers. The OPM also offered deferred resignation where federal employees could resign immediately and still be paid for the next several months. Meanwhile, those who decide to stay are not promised future employment and the memo stated new conditions for employees, that they be “loyal, trustworthy, and to strive for excellence in their daily work”; principles that likely will become benchmarks for future performance reviews.

Under the Trump administration, federal workforce reductions will happen, along with a greater deployment of artificial intelligence (AI), automation, and outsourcing to private firms. These new services will cost millions of dollars to design, deploy, and train the federal workforce, creating new national and data security threats as well, given the level of protected information at stake. But the influence of Big Tech leaders, who are formally and informally advising President Trump and his administration, may be accelerating a smaller government workforce based on their own values about corporate governance. Big Tech companies were among those that led the RTO mandates for their own employees after the pandemic with similar terms and conditions, as well as promises made that were not kept. Many of these same companies are making AI more technically advanced without realizing that millions of people are still impacted in the U.S. by the lack of digital access. As Biden era policies were working to address the connectivity challenges faced throughout the U.S., these programs are now being challenged, which will almost guarantee that even the best of AI technologies embedded in government functions may be inaccessible to most people.

Big Tech companies led the RTO movement

Seven months after his acquisition of Twitter, Elon Musk required an immediate RTO of employees, who were promised flexible work arrangements prior to the sale. Twitter was the first company to mandate such policies, resulting in the attrition of 80% of the workforce mainly through termination. Musk is now leading the newly formed Department of Government Efficiency (DOGE), leading the federal workforce reduction efforts, making the actions of the last few days eerily familiar. In 2023, Meta’s RTO mandate required that employees be in the office at least three days per week and threatened workers with termination if they did not oblige. Around the same time, Meta promoted its augmented and virtual reality (AR/VR) technologies, suggesting that hybrid employees could utilize the technology. In January, Amazon also joined its peers in requiring that employees be in the office five days per week, which already is leading to traffic congestion in its Seattle headquarters, lack of office space for returning employees, increased theft in open work environments, among other concerns being shared by employees.

The intent behind RTO policies to get more employees into the office is legitimate. Five years since the official end of the pandemic and employers are seeking to recreate collaboration and community among more isolated team members. Yet, private companies do not compare to the size of the U.S. federal government. The redundancy and minutia built into certain government jobs are intentional to ensure certain protections against fraud, theft, and most importantly, autocrats.

It’s also no secret that many Big Tech work environments offer other perks to employees to keep them in the office longer. From free meals and lunch break massages to exercise rooms, employees who have returned to the office benefit from private sector incentives, including pay. Instead of leaving children, elderly parents, and pets at home, some of these private technology companies allow their employees to bring two out of three loved ones to work by instituting on-site day cares, or “bring your pet to work” policies. These incentives will probably never make it into a federal office, especially as taxpayers foot the bill for government infrastructure.

In a 2005 blog on the importance of federalism, the late Brookings scholar, Pietro S. Nivola stated: “For the often indiscriminate preoccupation of national policymakers with the details of local administration is not just wasteful; it can be irresponsible.” But these efforts to reduce the federal workforce also come with actions to modernize it, which ultimately may result in privatization of government services through increasing automation of job roles that will be eviscerated and outsourcing of certain government functions.

Automation may come sooner than expected

In the same week of announcing federal furloughs, OpenAI introduced ChatGPT Gov, which will allow government agencies who were previously prohibited from using the large language model to access “non-public sensitive data.” In a press release, OpenAI discussed the future integration of AI into government services: “We believe the U.S. government’s adoption of artificial intelligence can boost efficiency and productivity and is crucial for maintaining and enhancing America’s global leadership in technology. This includes making our models available to support public sector work that benefits society–such as public health, energy and the environment, transportation and infrastructure, consumer protection, and national security.”

The federal government is no novice when it comes to the integration of AI into existing service delivery models. The Biden administration’s AI EO, recently rescinded by the current administration, directed government agencies to appoint a chief AI officer, identify plans for leveraging AI to improve government functions, and strategies for mitigating online biases, among other national security requirements in software applications. Surprisingly, Big Tech companies were intricately involved in designing this EO with particular emphasis on red-teaming to weed out technical vulnerabilities in applications. But with those AI mandates reversed, the federal government will become a Wild West for the experimentation, testing, and deployment of tools that may have consequential impacts on consumers. Further, settled procurement policies may be upended to get more products to the public sector market sooner.

For example, the Internal Revenue Service (IRS) attempted to revert to make their work more efficient by using AI systems to flag taxpayers for auditing. However, a 2024 independent report from the U.S. Government Accountability Office (GAO) found that claims for tax credits were disproportionately disqualifying eligible applicants. A year prior, a Stanford University study found that the IRS’s use of AI was more likely to flag Black taxpayers for unnecessary audits, despite their compliance with the tax code; the agency has since confirmed the report findings and stopped using the model.

The design and use of AI by federal agencies in lieu of human federal workers will likely lead to similar biases based on the data training these models, which may be over- or under-represented by populations who rely on government services. Without guidance on the privacy protections for consumers, along with some level of regulatory or legislative oversight, the federal government will become a minefield for exploitation, and in some instances, turn to outsourcing to third-party companies, who have their own precarious history harvesting enormous amounts of personal data.

Making the government vulnerable to an industry where data is consistently monetized for economic gain is, at best, a dangerous proposition to all consumers of government services where some communities will be harmed by biased training data, online mistakes in eligibility, and in many instances, the lack of basic internet access to benefit from increased amounts of digital services.

Getting most government services online forgets the digitally invisible

Across the U.S., many people lack digital access, especially those living in rural areas, older, or in areas where connectivity is unaffordable or not as robust when compared to more connected communities. Over the past five years, heightened public awareness of the realities of the digital disparities resulted in an increase in available public and private sector resources for states and municipalities to bridge the digital divide. Having internet access is a bipartisan concern because households must have the financial capacity to afford monthly internet services. Yet, advances in lowering the cost of broadband access were overturned when the Affordable Connectivity Program (ACP) ended in April of 2024, de-enrolling more than 20 million households who were benefitting from being connected after the pandemic. Further, many of these households were in rural, red states where guaranteed internet access finally didn’t compete with other pressing budgetary concerns, like food and transportation. As government leverages existing and emerging technologies to change the way work is performed, consumers will be on the receiving end, and if an individual or household lives in a digital desert, or has insufficient access, the translation and migration from in-person to online service delivery will be lost and potentially a waste of taxpayer dollars.

In 2011, my co-author, Jon Gant and I wrote an article about the levels of transparency and infrastructure needed to accomplish this massive feat. At that time, the Obama administration had moved toward an Open Government Directive to assemble more alignment in federal, state, and local resources. In addition to promoting more collaboration between the public and private sectors, one of the most important recommendations was to ensure that every day people could access these digital resources in their home or at local anchor institutions, including libraries, community organizations, and schools.

The point here is that no matter how much AI is embedded in improving and recalibrating traditionally analog positions, the consumer is at the center of the production cycle, and without internet access, AI and digital literacy, and greater transparency on how a more modernized government will work, efforts will be less effective. That is why more research and data are required when enacting such changes, and the influence of Big Tech companies, who are looking to government for more profit-incentive than public good, should be questioned.

The federal government is massive, and no one would disagree that opportunities to make it leaner are perhaps overdue. Public units need to become more efficient and effective. But changes embedded in the opaqueness of decision-makers, and driven by politically motivated issues are not the way to serve the American consumer of government services; too much is at stake, including taxpayer dollars. Further, policymakers need to be sensitive to the impact of possible changes on the existing workforce who represent the talent pool necessary to maintain order. Doing otherwise can result in widespread disruption, generating chaos and ineffectiveness.

  • Acknowledgements and disclosures

    Meta and Amazon are general, unrestricted donors to the Brookings Institution. The findings, interpretations, and conclusions posted in this piece are solely those of the authors and are not influenced by any donation.

The Brookings Institution is committed to quality, independence, and impact.
We are supported by a diverse array of funders. In line with our values and policies, each Brookings publication represents the sole views of its author(s).