Amazon Closes Mechanical Turk To New Customers
For ML and data teams, this is another signal that low-cost public crowd labeling is becoming legacy infrastructure rather than a default training-data pipeline. Amazon Mechanical Turk now says it will close to new customers on July 30, 2026, while existing users remain unaffected. TechCrunch reported the change on July 5, noting AWS described the decision as made after careful consideration and said it will keep investing in security and availability but does not plan new features. The MTurk site still points ML users toward SageMaker Ground Truth and Ground Truth Plus, making the practical migration path clearer: use managed labeling workflows, private or vendor workforces, or newer data operations stacks rather than opening a fresh MTurk requester account.
Practitioner Impact
Mechanical Turk becoming unavailable to new customers is not a model launch, but it matters to the data supply chain behind model training, evaluation, and human review. For teams that still treat public crowd work as an easy fallback for labeling, moderation, surveys, or validation, the announcement turns MTurk from an open onboarding path into legacy AWS infrastructure. The practical effect is narrower choice: new ML projects will need to plan around managed labeling services, private workforces, vendor workforces, or internal review operations from day one.
What Changed
The Amazon Mechanical Turk site now states that the service will be closed to new customers effective July 30, 2026, while existing users are not affected. The same page still describes MTurk as a crowdsourcing marketplace used for data validation, content moderation, research, and machine learning development, including collection and annotation workflows for training data.
TechCrunch reported the change on July 5 and said AWS described the decision as considered rather than abrupt. TechCrunch also noted that AWS plans to keep supporting security and availability for the service, but does not plan to add new features. That distinction is important: this is a maintenance move, not a full shutdown for current requesters or workers.
Why It Matters
MTurk has long occupied an awkward place in AI: it helped teams scale human annotation and evaluation, but it also carried quality, privacy, labor, and bot-risk concerns. AWS documentation for SageMaker Ground Truth and Amazon A2I already limits what can be sent to the Mechanical Turk workforce, including workloads with personal or protected health information. Closing new customer onboarding reinforces a broader shift from broad public microtask markets toward more controlled labeling and review pipelines.
For data-science teams, the operational lesson is simple: treat human-in-the-loop data work as a governed production dependency, not as an interchangeable crowd marketplace. New projects should budget for data quality, reviewer identity, privacy boundaries, adjudication, and audit trails before they need labels at scale.
Key Points
- 1Amazon Mechanical Turk says it will close to new customers on July 30, while existing users can continue operating.
- 2The change matters for ML teams that still use public crowd work for labeling, validation, or human review.
- 3Amazon is steering data-labeling buyers toward managed SageMaker Ground Truth workflows instead of new standalone MTurk onboarding.
Scoring Rationale
This is a solid practitioner-relevant infrastructure change for teams that still use MTurk for data labeling or human review. It is not a frontier-model or funding event, but it changes the onboarding path for a long-running service that has been part of ML data operations for years.
Sources
Public references used for this report.
Practice with real Retail & eCommerce data
90 SQL & Python problems · 15 industry datasets
250 free problems · No credit card
See all Retail & eCommerce problems
