Will AI upend or reinforce the neoliberal order?

Techno-utopians like OpenAI’s Sam Altman and Google’s Eric Schmidt are largely defining the U.S. conversation, making bold pronouncements that AI is poised to solve society’s biggest challenges. They have positioned themselves in favor of government oversight and regulation, provided they are at the table to design and benefit from them, ensuring they are not overly restrictive. At the other extreme, AI doomsdayers like Geoffrey Hinton, “the Godfather of AI,” and Elon Musk are raising alarm bells that AI poses an existential threat to the long-term future of humanity. Warding off this threat, in their telling, requires a different set of technologists to be in the driver’s seat of regulation and oversight.

Somewhere in the middle lie armies of AI ethics and safety personnel, with varying degrees of influence, and dozens of tech-accountability organizations developing policy frameworks and voluntary standards designed to guide how this new technology comes into society.

If this sounds like the backstory to a space opera that puts all our futures into the hands of a small set of tech sector actors, that’s because it is. This is taking shape by design, further cementing the tech’s oversight over its own regulation (see this New York Times op-ed and MIT Technology Review article for thorough overviews). The nature of the conversation, whether defined by doomsdayers and techno-utopians or the crowded space of AI ethics and integrity, can obfuscate the simple truth: The most important conversation we need to be having is not primarily about the newest game-changing technology, but about what kind of society we want to live in, who will decide how technology is developed and deployed, and how its productivity gains will be shared. This is ultimately a question of values, social power, and what kind of political economy will best serve humanity as neoliberalism loses its grip.

Over the last three decades, I have organized with people in low-wage jobs to advance shared prosperity, economic stability, and community accountability at some of the most powerful corporations in the world. Working closely with casino staff in Las Vegas, Silicon Valley janitors and security officers, and Walmart and Amazon workers around the country, I have seen first-hand how technology is reshaping work, the power of the technology sector, and the potential for organized workers and communities to build influence through collective action. The most exciting and effective campaigns succeeded by connecting workers’ core economic priorities with the role of their employers in shaping society and the economy, challenging their concentrated power.

Daron Acemoglu and Simon Johnson’s recent book, “Power and Progress,” provides a valuable framework to make meaning of the role of technology and the economy in this particular moment. They reorient the history of technology around questions of political economy, instead of assuming that progress is technology’s inherent end state. They argue that “a new, more inclusive vision of technology can emerge only if the basis of social power changes” and offer a road map based on the successes of the Progressive reform movement during the Gilded Age of the late-19th century, which laid the foundation for the New Deal social compact.

There are unmistakable parallels between the Gilded Age and the late-20th century’s Third Industrial Revolution, marked by new digital and computing technology and the rise of a new set of powerful corporations with powerful financial sector partners spanning venture capital and institutional investors. The ascendant tech sector grew up along with the neoliberal order, whose corporate and intellectual architects dismantled the New Deal social compact, prioritizing free-market capitalism, deregulation, and minimal government intervention. Silicon Valley emerged as the globe’s center of innovation and entrepreneurship, with people across the political spectrum embracing the notion that the heads of large technology corporations are better-positioned than governments to solve society’s problems and build a more prosperous world. As Gary Gerstle describes in “The Rise and Fall of the Neoliberal Order,” “The resulting union of hackers and capitalists would bring together left and right versions of neoliberal ideology. … The Silicon Valley story provides a window on many of the characteristics, anticipated and unanticipated, of the coming neoliberal order.”

Tech sector leaders championed individualism, free-market competition, and the idea that unfettered technological innovation could drive economic growth and societal progress, contributing to and embodying what became the neoliberal ethos. During this period, the sector propelled a systematic reframing of people’s social identities toward individual consumption, delivering products and services increasingly quickly and efficiently in exchange for total access to and ownership of people’s data and privacy. This tech-accelerated, neoliberal ideology also obscured the heavy government investment that provided the foundation for developing the internet, digital platforms, and computing.

The Big Tech titans (Facebook, Apple, Microsoft, Google, and, more recently, Amazon and Musk’s enterprises) have risen to unprecedented power and amassed Gilded-Age-level profits, competing as much with nation-states like China as with each other in setting the “rules of the road” for global communications, information, market, and data infrastructure. Through his Starlink network, Musk now owns a majority of the world’s satellite infrastructure, and Meta/Facebook owns, virtually without restriction, vast tracts of the digital commons that have replaced the role of traditional media, both with massive domestic and geopolitical consequences. Amazon, Google, and Microsoft dominate the cloud, which is the back-end of the private, public, and consumer sectors. The founders and owners of these companies are now the wealthiest human beings the globe has ever seen, with power and ego to match.

The human cost of the neoliberal order and the unaccountable dominance of Big Tech and venture capital are well-documented, severe, and well-known by readers of this newsletter. Reduced government regulations and oversight enabled Google, Microsoft, Amazon, and Facebook/Meta to consolidate into monopolies, impacting workers and consumers. Tech-driven algorithmic and data surveillance-based management, increasing automation layering onto economic dislocation from job displacement from de-industrialization, and the runaway inequality in wealth and opportunity are all serving to weaken economies, democracies, and societies around the globe.

As we stand at the precipice of a Fourth Industrial Revolution, with the launch of generative AI and the looming possibility of artificial general intelligence, another trajectory is both possible and necessary to ensure robust democracy and shared prosperity. However, society-wide engagement and the social power to advance an agenda for technology that is centered around public good are yet not in place, despite the civil society representation on President Biden’s National AI Advisory Committee, Senator Schumer’s Insight Forum, and groups developing voluntary guardrails and governance like Partnership on AI.

The AI startup landscape may have new names and faces, but new generative AI technology is being built directly on the old technology stack and integrated with existing products and services. As a result, productivity gains are set up to feed back to the investors in and owners of this infrastructure. As the ideologies and social order of the last few decades begin to come loose, these tech sector leaders are aggressively stepping into the public sphere and policy arenas, aiming to cement their power to set the terms of and be the beneficiaries of the new social order.

While jockeying for position among themselves and advancing different positions on regulation, it is clear that the industry is setting the context for these conversationsand advancing frameworks that protect their profits. They are fighting tooth and nail against those that threaten their core economic model. For example, tech firms are pouring millions into lobbying and contributions to potential civil society critics. Members of the effective altruism movement, backed by tech-generated fortunes, are working in the halls of Congress to reframe policy around long-term, existential risks instead of current and imminent harms. Google is leaning into healthcare regulation that will cement its market capture, and Meta is challenging the constitutionality of the FTC’s antitrust mechanisms.

We need to build and wield what Acemoglu and Johnson call “countervailing force” to redirect technology for public good and purpose. We have seen how unions can set standards and guardrails over technology and how labor actions can shift society’s understanding of what is at stake. For example: UNITE HERE has established collective bargaining language over third-party technologies; the Writer’s Guild went on strike to protect people’s rights to their creative products; and the NFL Players Association won contract language that restricted the league’s use of players’ biometric data. We have also seen significant action by communities campaigning for Google and Amazon to stop selling their technologies for immigration enforcement, criminal justice, and militaries; legal action to challenge working conditions in the AI supply chain; and campaigns challenging racially discriminatory algorithms at Microsoft, Google, and Facebook/Meta, as well as Amazon’s monopolistic actions. It is critical to expand the involvement of communities and workers most directly impacted by AI and technology in its governance.

Generative AI will be integrated into workplaces and coexist with algorithmic management and surveillance technology. As key foundational steps, we must establish clear policies and regulations that ban or limit the use of particularly extractive and harmful technologies, expand people’s right to organize, and build the capacity of workers to bargain and engage over their introduction and use. As generative AI is poised to significantly increase the efficiency of customer service, coding, and other jobs, the AFL-CIO’s Technology Institute, UNI Global, CWA, and others rightly call for this technology to augment, not replace, existing workers. As this happens, we need to understand the productivity gains resulting from generative AI and make claims on distributing those gains to workers and communities.

Organizing models

People working directly in the tech sector, whether they are Silicon Valley scientists or AI supply chain workers in the Global South, play a crucial role in shaping the direction of tech. They have deep knowledge about their industry and, often, values and interests distinct from the sector’s business and financial leadership. Big Tech defectors, including those who have built AI Now, TechEquity Collaborative, and the Integrity Institute, have already reshaped the public conversation, providing thought leadership and policy frameworks from an insider’s perspective. Campaigns run by workers in the economically precarious and largely invisible supply chain have begun to expose the human labor required to build and sustain AI and digital platforms, as well as the steep human cost. The model of “regulation from below,” where unions and worker organizations play a watchdog role over the practices of their employers, is an exciting model to explore and apply. Amazon Employees for Climate Justice is an example of this model in action, organizing as shareholders and through walkouts to pressure Amazon to commit to reducing their carbon footprint.

Bargaining for the Common Good, a framework where workers organize, campaign, and bargain for broad common good goals, such as reduced class sizes or expanded public benefits, is another model that can apply to AI’s incorporation into public institutions. Public sector workers could call for state and local governments to use AI and algorithms to identify people who are eligible for but not receiving public benefits, and draw bright lines against automated systems removing people from those services. How people interface with public institutions — health care systems, public services, K-12, higher education — is increasingly mediated through technology. Expanding the common good framework to include worker and community governance infrastructure can provide vehicles for people to shape the use of technologies that most impact their lives.

The Athena Coalition, which explores and challenges the multiple impacts and harms that Amazon’s concentrated power has on society, is another model that brings workers and impacted communities into shared strategy and campaigning. Athena connects workers challenging algorithmic workplace management with small businesses being squeezed by Amazon’s monopolistic practices and employees at Amazon headquarters challenging their military contracts. Developing cross-issue coalitions and ecosystems like Athena to engage in this moment is a critical counterweight to the technology sector’s power.

Our present-day choices can shape a new post-neoliberal social compact, strengthening democracy and advancing shared prosperity. As science fiction writer Ted Chiang shares in his incisive New Yorker op-ed, the greatest risk of AI is less about humans losing control of supercharged AI systems and more about “supercharged corporations destroying the environment and the working class in their pursuit of shareholder value.”