Amazon will stop accepting new customers for Mechanical Turk

Amazon is closing its landmark Mechanical Turk crowdsourcing platform to new customers beginning July 30, 2026, signaling the end of an era for a foundational service in the AI ecosystem. The announcement from Amazon Web Services (AWS) confirms that while existing customers can continue using the platform, development on new features has ceased. This decision effectively puts the nearly two-decade-old service, once a critical source for human-powered data annotation, on life support as the industry shifts toward more automated methods.

A Slow Sunset for a Pioneering Platform

• New Customer Cutoff: The service will be closed to new requesters starting July 30, 2026.
• Maintenance Mode: AWS will continue to support security and availability for existing customers but plans no new feature development.
• Legacy Impact: Launched in 2005, the platform was crucial for training early neural networks and was later integrated into Amazon's SageMaker AI service.

Mechanical Turk's relevance has waned as the very technology it helped build has surpassed its utility. The platform became mired in a unique paradox, with a 2023 analysis finding that up to 46% of its human workers were using large language models to complete their tasks, severely degrading the quality and reliability of its data. For years, users have also reported a decline in the platform’s quality due to bots and fraud, leading many researchers and companies to seek alternatives for high-stakes data labeling.

The winding down of Mechanical Turk highlights a major evolution in the AI data pipeline. The industry is moving away from the brute-force, micro-task model that defined the platform's heyday. Instead, companies are increasingly turning to automated labeling, synthetic data generation, and more specialized data services that offer higher quality and consistency. The platform's fate underscores how the economic model for massive, low-skill human annotation workforces is becoming obsolete in the age of advanced AI.

The slow demise of Mechanical Turk isn't just the end of a product; it's a market signal that the AI industry's reliance on brute-force, manual data annotation is waning. As AI models become capable of generating and validating data themselves, the economic viability of massive, low-skill human workforces for simple labeling tasks is fundamentally breaking down, pushing the ecosystem toward synthetic data and more sophisticated human-in-the-loop processes.