How AI Is Turning Full-Time Jobs Into Gig Work
Source: To Vima / The Guardian
Published: 2026-06-24
Entity Analyzed: Gigification of Knowledge Work
URL SCAN
In 2024, buy-now-pay-later company Klarna announced it would cut hundreds of customer service positions and replace them with an AI chatbot. A year later, after customers complained about declining service quality, Klarna quietly began rehiring human customer service agents. At first glance, this seemed like a win for workers in the age of AI. The reality, however, is more complicated. Rather than bringing back full-time employees, Klarna hired workers under what its CEO described as an ‘Uber-style model.’
The Triage
This is not a story about AI replacing jobs. It is a story about AI disassembling jobs. The Klarna case is the purest form of the new model: the AI chatbot does not replace the worker—it replaces the employment relationship. The full-time job is not automated; it is fractured. The routine portion goes to the machine. The complex portion goes to a gig worker with no benefits, no protections, and no guarantee of hours. The CEO’s ‘Uber-style’ comment is not a slip—it is the business model. The same logic that turned car ownership into on-demand labor is now turning knowledge work into on-demand fragments. Customer service, copywriting, financial analysis, legal assistance, software development, nursing—these are not being eliminated. They are being decomposed into task fragments and reassembled as gig work. The article documents this as a trend. The Oracle sees it as the operating system of the new economy.
The Autopsy (with DT-LAG)
Mechanical Collapse Point
The structural shift is not driven by AI capability but by classification. The moment a company can legally reclassify a worker as an independent contractor, the economics invert. The employer pays no benefits, no payroll taxes, no unemployment insurance, no workers’ compensation, no overtime, no sick leave. The 60 million Americans already in freelance/contract work (39% of the workforce) represent not a labor market but a reclassification market. The projection to 86 million by 2027 is not about worker choice—it is about employer optimization. The nurse platforms (ShiftMed, CareRev, Clipboard Health) are not ‘innovation in healthcare staffing.’ They are labor arbitrage engines operating at the boundary of regulatory classification. The 17 states that recognize them as ‘healthcare worker platforms’ rather than staffing agencies have not merely deregulated; they have decoupled healthcare labor from labor protections. The creative professionals training the AI that replaces them—algorithmic composers, AI text raters, actors reducing background cast needs—are not the exception. They are the template: the worker who funds their own obsolescence one gig at a time.
Lag-Weighted Social Timeline
The lag here is 6-18 months, not years. The legal framework for classification is already shifting. The political recognition that ‘gig’ means ‘unprotected’ is not yet widespread, but the economic reality is already priced in. The 39% figure is a lag indicator: the policy will follow the money. The Klarna model is the first wave. The next wave is every company with a customer service department, a content team, a finance operation, a legal review function. The lag is not technological—it is regulatory theater.
Lag Factors
Classification Inertia: The legal distinction between ’employee’ and ‘independent contractor’ is being exploited faster than legislatures can redefine it. The Department of Labor’s 2024 rule is already being challenged and will be litigated for years while the reclassification continues.
Cultural Coping: The ‘freelancer’ and ‘digital nomad’ branding masks the precarity. Workers are told they are ‘entrepreneurs’ and ‘their own boss’ while algorithms assign tasks, set rates, and evaluate performance. The language of autonomy is a lag factor.
Union Response Time: The Kaiser strike and UC unionization are early indicators, but union density in the US is below 10%. The leverage is gone before the organization forms. The platforms that control the labor have already priced the union premium out of the model.
Physical World Inertia: The ‘Uber for nursing’ platforms are already operating in 17+ states with billion-dollar valuations. The regulatory capture is not pending; it is complete. The inertia is not in the system—it is in the public’s understanding of what has already happened.
Defensive Moats
Regulatory Armor: Licensing and certification requirements in healthcare, law, and finance create barriers to full gigification—but the platforms are already finding workarounds (nursing platforms operate inside regulatory gray zones).
Trust Shield: The ‘human touch’ argument is still deployed in customer service and healthcare, but Klarna’s quality complaints and rehiring of gig workers (not full-time) shows the trust shield is being bypassed, not honored.
Physical Chains: Geographic licensing and local labor laws are the last line of defense—but the ‘platform’ model is designed to route around them by operating as marketplaces rather than employers.
The moats are regulatory, not technological. And regulation is losing the race.
Future-Proofing Scorecard
| Timeline | Score | Commentary |
|———-|——-|————|
| 1 year | 2/10 | The reclassification wave is already 39% of the workforce. The infrastructure is built. The next 12 months will see acceleration, not awareness. |
| 2 years | 1/10 | Knowledge work gigification becomes the default. ‘Full-time’ becomes a niche luxury for senior architects and executives. The protections built over generations are gone for the majority. |
| 5 years | 0/10 | The employment contract as a social institution is replaced by task-market platforms. Benefits, tenure, and labor protections are historical artifacts. |
| 10 years | 0/10 | The concept of ’employee’ is as anachronistic as ‘factory hand’ in a post-industrial economy. The platform owns the task, not the worker. |
The Verdict
This article is a chronicle of the job being dismantled, not displaced. AI does not need to outperform a full-time worker to destroy the full-time job. It only needs to outperform the routine portion of the job, leaving the complex fragments to be sold as gigs. The Klarna model—AI + Uber-style labor—is the blueprint. The creative professional training their replacement is the emotional core. The nurse working through a gig app without stethoscopes or protections is the physical reality. The 39% figure is not a statistic about freelancing; it is a measure of how much of the workforce has already been decoupled from the protections that made the modern economy survivable. The verdict: this is not the future of work. It is the present of work, and the past of employment. The job, as a social institution, is being unbundled. The platform does not employ. It extracts.