AI Is Contributing to the Gigification of Work

AI hype is being used by companies as a cover to replace stable employment with precarious gig work, according to research and interviews. The trend accelerates gigification as laid-off tech workers turn to freelance platforms like Upwork, while many jobs created to fix AI-generated code are themselves gig-based rather than full-time positions.

AI Is Contributing to the Gigification of Work Bosses have desired ways to cut labor costs since time immemorial. Artificial-intelligence hype provides a powerful new excuse to replace stable employment with gig work. “AI Code = Dumpster Fire https://www.linkedin.com/posts/coderabbitai nyc-takeover-activity-7425216406373322752-fvqu ?” asks an advertisement wrapped around a set of pillars in New York’s Penn Station. “We can help,” it answers. The ad is for CodeRabbit, a company whose software identifies and fixes bugs in code written by artificial-intelligence software. Besides the obvious irony of such a product — if AI can replace software engineers, why does AI-generated code require another AI to clean up its mistakes? — the ad conceals a more important reality: much of the work created to fix AI slop is done not by AI but by a new class of gig workers. As Devvin, a tech consultancy owner, told me in a recent interview, the boom in AI products available to the public meant that start-up founders without a computer science background could now “vibe code,” using AI chatbots to create unpolished but functional app prototypes. Yet when it came to scaling the app to meet the needs of customers and investors, they needed help, usually quickly and cheaply. “When they have a problem,” Devvin explains of new founders, “they’re not going to go to Google and search for a legitimate software company in my area; they’re gonna go to Upwork an online platform that connects freelance engineers with customers for one-off gigs and search for software developers.” Contracting and freelance work in the computing industry are not new. Subcontracting in the tech industry led to the unionization of contingent tech workers at Microsoft in the late 1990s. Freelancers or gig workers on platforms like Upwork and Fiverr live precarious lives, waiting for gigs that they don’t know will come. Whole subreddits https://www.reddit.com/r/UpworkOfficial/comments/1ucfdei/got laid off recently thinking of becoming a/ are dedicated to Upwork contractors recounting months without work and their struggle with the lack of stable income that comes with contracting. AI is not eliminating jobs because it performs so well. Instead, the hype surrounding “AI” conceals the continued growth of businesses reliant on precarious workers. While popular fears about the impact of technology center on mass unemployment, what is happening is perhaps worse. The story isn’t that AI is leading to mass unemployment; it’s that AI is being used as a cover for the gigification of work. The Gigification of Work Platformization and misclassification have been the story of work for the past few decades, impacting not only Uber drivers and Instacart delivery workers but also workers who have or had protected status as employees. As I have found in my research https://urldefense.com/v3/ https:/www.tandfonline.com/eprint/GF98CJBNYGMNY9DTBBHJ/full?target=10.1080 00380253.2026.2639126 ;Lw BDUfV1Et5lrpZQ QvYwb3TaLWzz5nWTMh0CK0LviNPRUAs5KeiWA7r4Q2EpRA1m 5r8 5FVkibaV5DKUnoAyc02fu34ydr75gc8o7c$ on the impact of gig work on the retail industry, companies that employ traditional employees have worked hard to use technological change as an excuse to transform employees into contractors. The hype around AI speeds up gigification by creating a nebulous narrative that allows companies to push full-time employees out. This increases the reserve army of labor, leaving displaced workers more willing to take worse, gig-work versions of the jobs they once held — or whatever gigified work is available elsewhere. The mass layoffs under the cover of AI in “cushy” tech companies like Amazon and Google over the past few years are pushing these workers into contracting jobs. At the same time, many of the jobs created to fix AI’s problems are not full-time, stable jobs but precarious gig work. Laid-off workers can access some help from the anemic welfare state, such as unemployment insurance, though the generosity of those benefits varies from state to state. For those in the platform economy, however, their status as contractors means they are excluded from unemployment insurance and from the protections of employment such as minimum wage, workers compensation, paid leave, and health insurance. Gigification is the latest step in a decades-long trend in the United States and around the world: the hollowing out of permanent, full-time work and its replacement with piece-rate work. Platforms mediate this work through apps while classifying workers as independent contractors, allowing companies to evade many of their employer obligations. The legal battle over the classification of gig workers has played out across federal, state, and municipal arenas, including through back-and-forth legislative fights in California about whether rideshare drivers are workers or contactors. Given the opportunity, most companies will eagerly lay people off and rehire them as contractors for cheaper. There are websites and law firms dedicated to https://www.hireborderless.com/post/rehiring-an-employee-as-a-contractor-what-you-should-know instructing companies on how to get around labor laws about misclassification. But the rise of platforms has accelerated that dynamic fantastically. To really understand the impact of platform firms on workers, we must look beyond these platforms themselves and to their broader impact on society. The Appification of Vocation These companies aren’t just structuring the lives of thousands of workers who contract for them; they are reshaping the landscape of work across industries. In my ethnographic research https://urldefense.com/v3/ https:/www.tandfonline.com/eprint/GF98CJBNYGMNY9DTBBHJ/full?target=10.1080 00380253.2026.2639126 ;Lw BDUfV1Et5lrpZQ QvYwb3TaLWzz5nWTMh0CK0LviNPRUAs5KeiWA7r4Q2EpRA1m 5r8 5FVkibaV5DKUnoAyc02fu34ydr75gc8o7c$ on grocery workers in California, I found that the partnerships and overlaps between platform companies like Instacart and DoorDash and grocery stores helped to erode the employment status and security of grocery workers, even those who were permanent employees with a relatively strong union to protect them. Grocery companies’ partnerships with these platforms provided them with a cheap way to reduce their labor costs. They could replace their own e-commerce, or online ordering, workers because many of the tasks — preparing online orders and delivering groceries — were the same. Even in cases where they retained employees or hired more, they reorganized the work around technologies modeled on companies like Instacart, making the jobs resemble gig work. This meant increased surveillance, speedup, and overwork, making employees even more vulnerable to the whims of managers who determined their schedules. In the grocery industry, gigification exacerbates trends of precarity. The industry, through major consolidations, mergers, and acquisitions by private equity firms, has come to epitomize work in the low-wage service sector. What was once a lifetime career is now a transient, part-time job. But gigification is a problem beyond the grocery industry. Subway ads in New York City market a platform for home care workers, where elderly and disabled patients can be matched with a different carer every day. These ever-changing, contingent care workers will have no knowledge of patient health, personality, needs, or routines. And these workers are not even considered employees covered by minimum wage and health and safety laws, much able to join a union. Increasingly, nurses https://ainowinstitute.org/publications/uber-for-nursing also face gigification. AI models can’t yet replace nurses, but they do determine wages and schedules through platforms. A depressing symptom of both the academic job market’s collapse and the declining security of high-skilled jobs is a blog called “After Your PhD LLC” that instructs PhD and MA recipients on how to get jobs as data labelers for AI https://afteryourphd.com/phd-job-listings/data-labeling-ai-training-phds/ . The proximity of platform companies to traditional firms has spillover effects that increase gigification. The growth of AI is having a similar effect, though through a different mechanism: that of obfuscation. The hype around AI allows companies to lay workers off under the guise of technological advancement. Whether in the university, the tech firm, the grocery store or hospital, employers are searching for ways to evade the costs of traditional employment. Whatever labels academics give it — fissuring, precarity, underemployment, or a litany of other terms — the transformation is the same: employers are converting workers into contractors with no rights or protections. The Tactical Use of Hype In an economy where value is divorced from reality, speculation and hype become forces sufficient to reshape the economy. It doesn’t matter if AI is good enough to replace workers; it only matters that we believe it could be. The financial incentives to replace workers with AI are mixed. Companies have until now been buying AI “tokens,” or units of https://www.revenera.com/blog/software-monetization/ai-tokens-brief-guide-software-monetization/ AI capability, at a heavy discount, and recent headlines https://www.forbes.com/sites/jemmagreen/2026/07/02/ai-costs-more-than-the-people-it-replaced/ indicate companies are grappling with the high cost of it at full price. Its impacts on productivity are similarly dubious. One National Bureau of Economic Research paper finds https://www.thehrdigest.com/what-is-ai-washing-and-why-has-it-been-linked-to-layoffs/ that 80 percent of firms in the United States, UK, Germany, and Australia reported no impact of AI on productivity, and a Massachusetts Institute of Technology report https://mlq.ai/media/quarterly decks/v0.1 State of AI in Business 2025 Report.pdf indicated that 95 percent of AI pilot programs did not deliver substantial results. But it doesn’t matter if AI can replace workers. In the short term, whether it increases productivity or just produces slop is beside the point. Companies have always looked for an excuse to cut costs, and here AI presents itself to them. To pay for the cost of AI contracts, they can lay off workers they wanted to lay off anyway; they can use contractors instead of employees; they can make fewer workers do more. Companies can claim productivity gains https://fortune.com/2026/05/11/ai-automation-layoffs-gartner-study-roi/ even when those gains come from lowering employee head counts. We have seen this pattern before. “Driverless” Waymo cars are actually monitored https://www.automotiveworld.com/news/waymo-defends-overseas-remote-staff-in-us-robotaxi-ops/ by workers in the Philippines. Amazon Go stores, where customers could simply walk out with their purchases, relied on workers in India https://www.businessinsider.com/amazons-just-walk-out-actually-1-000-people-in-india-2024-4 who manually scanned surveillance data to see which items they were buying. Self-checkouts have created a whole new host of human jobs dealing with the customer frustrations of the faulty machines. It is difficult to not notice that much of the hidden labor behind so-called smart technologies is performed by the lowest-paid workers, often in other countries. Rather than replace human labor, these technologies often downgrade and relocate it. They transform it into a lower-waged, often “lower-skilled” job, increasingly displaced to the Global South, where “AI” is being touted as a new development strategy by the World Bank https://live.worldbank.org/en/event/2025/artificial-intelligence-foundations-agriculture-from-farms-to-future-economies . This is not to say AI models can’t be useful for anything or that they can’t or won’t be able to replace any work. Rather, it is to highlight the role of firms in seizing moments of change, crisis, or confusion to enact what is one of the most pervasive trends of our time: pushing workers into gig-like arrangements. Seen in this light, AI is not just a technological development — it is also a social process. What Is to be Done? In 1963, writing about the fate of American, particularly African American, autoworkers, James Boggs lamented https://www.historyisaweapon.com/defcon1/amreboggs.html “the new post 1945–46 vicious circle in which wage increases and fringe benefits would be won by the union and hailed as great social progress, only to be followed by concessions of some part of the control over production which the workers had won.” Though Boggs wrongly predicted that automation would shortly lead to mass unemployment, the issues he raised were pertinent. Responding to automation throughout the mid-twentieth century, American unions largely turned away from bargaining for control over production, instead becoming embedded within the company bureaucracy and settling for bargaining over the distribution of wages and benefits. This consensus has remained, with broad management rights clauses in most union contracts that limit workers’ abilities to contest decisions about the very work they do. AI provides an opportunity to revisit this consensus. The Writers Guild of America’s 2023 strike https://apnews.com/article/hollywood-ai-strike-wga-artificial-intelligence-39ab72582c3a15f77510c9c30a45ffc8 focused largely on limiting the ways that managers could deploy AI, protecting writers from having their work recycled or replaced by AI. Interestingly, tech workers, who have historically enjoyed high wages, haven’t had to bargain over wages and benefits and have primarily organized around control over the job — over what they produce. Perhaps because they are newcomers to the union movement, they have not been bogged down with this history of what workers should be organizing about. To deal with AI, unions must be creative and precise. Worker inquiries https://ai-workers-inquiry.github.io/ , a tool for building knowledge about specific working conditions, for example, would allow unions to map the specific ways AI is impacting the workplace. Doing this will help to break down the threat of AI into specific cases and violations that can be addressed in effective ways. For example, I found in my research that grocery store workers repeatedly had a hard time securing more hours at work. However, because their day jobs consisted of the same tasks as Instacart workers picking items off shelves and bagging them , many of them took shifts from Instacart and DoorDash in their spare time — often shopping at the same stores they worked at. They performed the same tasks at their same worksite but as contractors, not employees. Those hours did not count toward full-time status, and they were not covered by the union or minimum wage laws. If this kind of slippage can happen in a unionized environment where workers are protected by a contract, the opportunities for employer malfeasance is tremendous. Instead of treating AI as an inevitable force, the labor movement ought to identify the specific ways in which employers can use it to intensify work, erode protections, and increase gig work. This means challenging gigification at the bargaining table and through legislation. The excuse that AI gives to bosses to reorganize work is simply a narrative. It is open to contestation. Countering this narrative will help the American labor movement to reclaim the strength it needs.