Amazon closes Mechanical Turk to new customers, signaling end of an era for crowdsourced labor

Amazon will stop accepting new customers for Mechanical Turk on July 30, 2026, effectively freezing the growth of a pioneering crowdsourcing platform that has been at the center of debates about labor ethics, AI development, and the gig economy for two decades. The announcement, posted on the Mechanical Turk website, states that existing customers can continue using the service, but Amazon Web Services (AWS) will not introduce new features. The company described the decision as the result of “careful consideration.”

What the announcement means

While the service is not being shut down entirely, the decision to close registration for new customers marks a significant retreat. Existing requesters can still post tasks, and workers — known as “Turkers” — can continue completing them. But without new customers, the platform’s ecosystem is unlikely to grow. AWS said it will continue to invest in security and availability improvements, but the lack of new feature development suggests Mechanical Turk is now in maintenance mode.

The move follows years of declining usage and mounting criticism. One Reddit user summed up the sentiment of many in the community: “Someone at Amazon is going to decide keeping the Mturk servers running is a waste of time and resources and pull the plug entirely.”

A brief history of Mechanical Turk

Launched in 2005, Mechanical Turk was designed as a marketplace for “human intelligence tasks” — small, repetitive jobs that computers struggled to perform. Workers were paid tiny sums to complete tasks like identifying objects in photos, transcribing audio snippets, or determining the sentiment of a sentence. The platform’s name was a deliberate nod to a famous 18th-century hoax: a chess-playing “automaton” that actually concealed a human operator.

At its peak, Mechanical Turk was a vital tool for researchers, startups, and large companies needing human judgment at scale. It also became a flashpoint in discussions about fair wages, worker rights, and the hidden human labor behind automation. In the early 2010s, it was revealed that data from the platform had been used in the early stages of the Facebook-Cambridge Analytica scandal, further tarnishing its reputation.

The AI paradox: bots and language models

In a twist of irony, the same AI advancements that Mechanical Turk helped train have now made the platform less relevant. Starting in 2018, Amazon began positioning Mechanical Turk as a tool for annotating training data for machine learning models, integrating it with its SageMaker AI service. But by 2023, a study found that between 33% and 46% of Turkers were using large language models to complete their tasks, raising questions about the quality of the data and whether human involvement was still necessary.

This “snake eating its own tail” dynamic has accelerated the platform’s decline. As AI models improve, the need for cheap human labor for simple tasks diminishes. At the same time, the presence of bots and AI-generated responses has eroded trust in the platform’s output, making it less attractive for researchers and companies that rely on high-quality human annotations.

What this means for workers and researchers

For the thousands of workers who have relied on Mechanical Turk for supplemental income, the announcement is another blow to an already shrinking gig economy niche. Many have reported declining earnings and increasing competition from bots in recent years. For academic researchers, the platform was once a go-to source for behavioral studies and data collection, but concerns about data quality and worker attrition have already driven many to alternatives like Prolific or Cloud Research.

The broader implication is that the era of “human-in-the-loop” crowdsourcing for AI training may be nearing its end, at least in its current form. As AI models become more capable, the demand for cheap, repetitive human labor is likely to shift toward more specialized, higher-skill roles — or disappear entirely.

Conclusion

Amazon’s decision to close Mechanical Turk to new customers is a quiet but significant milestone. It marks the end of a chapter in which crowdsourced human labor was seen as a necessary complement to automation. The platform that once symbolized both the promise and the peril of the gig economy is now being overtaken by the very technology it helped create. While the service still exists for now, its future looks increasingly uncertain.

FAQs

Q1: Can I still sign up as a new Mechanical Turk worker?
Yes, the announcement only applies to new customers (requesters). New workers can still register and complete tasks, but the lack of new customers means fewer tasks may be available over time.

Q2: Will existing Mechanical Turk requesters be affected?
No. Existing customers can continue to use the service as before. AWS says it will maintain security and availability, but no new features will be added.

Q3: Why is Amazon making this change now?
Amazon has not provided a detailed explanation, but the decision likely reflects the platform’s declining relevance due to AI advancements, increased fraud and bot activity, and shifting priorities within AWS.