Amazon Augmented AI

aid: amazon-augmented-ai
name: Amazon Augmented AI
description: Amazon Augmented AI (Amazon A2I) is a machine learning service that makes it easy to build the workflows required
  for human review of ML predictions. Amazon A2I brings human review to all developers, removing the undifferentiated heavy
  lifting associated with building human review systems or managing large numbers of human reviewers.
type: Index
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
tags:
- Amazon Augmented AI
- Human In The Loop
- Machine Learning
- AI Review
- AWS
url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/apis.yml
created: '2026-03-16'
modified: '2026-06-20'
specificationVersion: '0.19'
apis:
- aid: amazon-augmented-ai:amazon-augmented-ai-api
  name: Amazon Augmented AI API
  description: REST runtime API for Amazon Augmented AI covering human loops for human review of machine learning predictions.
  image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
  humanURL: https://docs.aws.amazon.com/augmented-ai/2019-11-07/APIReference/Welcome.html
  baseURL: https://a2i-runtime.sagemaker.us-east-1.amazonaws.com
  tags:
  - Amazon Augmented AI
  - Human In The Loop
  - Machine Learning
  - AI Review
  - AWS
  properties:
  - type: OpenAPI
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/openapi/amazon-augmented-ai-openapi.yml
  - type: SpectralRules
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/rules/amazon-augmented-ai-spectral-rules.yml
  - type: Vocabulary
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/vocabulary/amazon-augmented-ai-vocabulary.yaml
  - type: JSONSchema
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/json-schema/a2i-human-loop-summary-schema.json
  - type: JSONSchema
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/json-schema/a2i-start-human-loop-request-schema.json
  - type: JSONStructure
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/json-structure/a2i-human-loop-summary-structure.json
  - type: JSONStructure
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/json-structure/a2i-describe-human-loop-response-structure.json
  - type: JSONLD
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/json-ld/amazon-augmented-ai-context.jsonld
  - type: Example
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/examples/a2i-start-human-loop-request-example.json
  - type: Example
    url: https://raw.githubusercontent.com/api-evangelist/amazon-augmented-ai/refs/heads/main/examples/a2i-describe-human-loop-response-example.json
  - type: Overlay
    url: overlays/amazon-augmented-ai-openapi-overlay.yaml
  - type: MCPServer
    url: mcp/amazon-augmented-ai-mcp.yml
  - type: ErrorCatalog
    url: errors/amazon-augmented-ai-problem-types.yml
  - type: Conformance
    url: conformance/amazon-augmented-ai-conformance.yml
  - type: Lifecycle
    url: lifecycle/amazon-augmented-ai-lifecycle.yml
common:
- type: VulnerabilityDisclosure
  url: security/amazon-augmented-ai-vulnerability-disclosure.yml
- type: DomainSecurity
  url: security/amazon-augmented-ai-domain-security.yml
- type: Authentication
  url: authentication/amazon-augmented-ai-authentication.yml
- type: Packages
  url: packages/amazon-augmented-ai-packages.yml
- type: LLMsTxt
  url: llms/amazon-augmented-ai-llms.txt
- type: WellKnown
  url: well-known/amazon-augmented-ai-well-known.yml
- type: Features
  data:
  - Human review integration for Amazon Rekognition and Amazon Textract
  - Custom flow definitions for any ML use case
  - Built-in worker task templates for common review tasks
  - Integration with Amazon SageMaker Ground Truth for workforce management
  - Private, vendor, and Amazon Mechanical Turk workforce support
  - Automatic routing based on ML confidence scores
  - Audit trail with evidence of human review decisions
  - Scalable workforce management across thousands of reviewers
  - Pre-built UI templates for image and text review tasks
  - Compliance support with PII content classifiers
- type: UseCases
  data:
  - Review low-confidence document text extraction results
  - Validate image classification predictions before deployment
  - Moderate user-generated content with human reviewers
  - Ensure accuracy of medical record processing
  - Verify identity document data extraction results
  - Build training datasets with human-verified labels
- type: Integrations
  data:
  - Amazon SageMaker
  - Amazon Rekognition
  - Amazon Textract
  - Amazon S3
  - Amazon SageMaker Ground Truth
  - Amazon Mechanical Turk
  - AWS IAM
  - Amazon CloudWatch
  - AWS Lambda
  - Amazon SNS
maintainers:
- FN: Kin Lane
  email: kin@apievangelist.com