Amazon shuts down Mechanical Turk as AI data labeling market evolves
Amazon is closing its legacy Mechanical Turk platform next month, highlighting a structural shift in the enterprise data labeling market as specialized AI startups capture share and corporate clients scramble for alternatives.
Amazon will shut down its AWS Mechanical Turk platform effective September 30, 2026. The company announced the decision Tuesday, stating it regularly evaluates its services and makes adjustments based on those assessments.
Launched in 2005 to help label data for Amazon’s webstore, the crowdsourced platform matched workers with simple digital jobs known as Human Intelligence Tasks. Founder Jeff Bezos famously described the service as "artificial artificial intelligence" because it farmed out tasks that were too challenging for computers at the time.
The closure highlights a rapid transformation in the AI data supply chain. While Mechanical Turk once served more than 500,000 workers, the platform has been in decline as advanced AI models and specialized data startups like Scale AI, Mercor, and Prolific have entered the market.
Amazon had recently attempted to reposition Mechanical Turk as a data annotation source for its SageMaker service, promoting its continuous workforce for fast turnaround times. However, the platform faced growing scrutiny over data quality, including a 2023 Swiss academic study revealing that up to 46 percent of workers used AI models to complete their tasks.
The wind-down is already disrupting enterprise clients. According to Krista Pawloski, a data worker and organizer with the advocacy group Turkopticon, some insurance and travel companies still dependent on the service are now scrambling to find alternative platforms. Amazon stopped accepting new Mechanical Turk customers last month, foreshadowing the discontinuation.
The shutdown also threatens a legacy gig workforce that relied on the platform for flexible, remote income. Pawloski, who began using the service in 2008 and later organized workers through Turkopticon, noted that many individuals still perform the work full time and are concerned about the transition.
As Amazon redirects resources away from legacy crowdsourcing operations, the move underscores a broader market reality. Enterprises can no longer rely on outdated micro-task models to fuel modern artificial intelligence development.
