Robust Intelligence

aid: robust-intelligence
url: https://raw.githubusercontent.com/api-evangelist/robust-intelligence/refs/heads/main/apis.yml
name: Robust Intelligence
type: Index
image: https://kinlane-productions.s3.amazonaws.com/apis-json/apis-json-logo.jpg
tags:
- AI Security
- Runtime Protection
- Algorithmic Red Teaming
- LLM Security
- Cisco
- Acquired
- Guardrails
description: Robust Intelligence is an AI security company founded in 2019 to defend ML and GenAI systems against adversarial
  attacks, data poisoning, prompt injection, and unsafe outputs. Its platform combined automated red teaming (Algorithmic
  AI Red Teaming) with runtime protection (AI Firewall) for LLM and traditional ML applications. Cisco announced the acquisition
  of Robust Intelligence in August 2024 and the technology now powers the Cisco AI Defense product line, with Robust Intelligence's
  capabilities integrated into Cisco's broader security portfolio.
created: '2026-05-23'
modified: '2026-05-23'
specificationVersion: '0.19'
apis:
- aid: robust-intelligence:cisco-ai-defense
  name: Cisco AI Defense
  tags:
  - AI Security
  - Runtime Protection
  - LLM Security
  - Cisco
  humanURL: https://www.cisco.com/site/us/en/products/security/ai-defense/index.html
  properties:
  - url: https://www.cisco.com/site/us/en/products/security/ai-defense/index.html
    type: Documentation
  description: Cisco AI Defense is the post-acquisition home of Robust Intelligence's AI security technology. It provides
    runtime protection for AI applications, model validation, algorithmic red teaming, and visibility into shadow AI usage
    across the enterprise. Delivered through Cisco's commercial sales channels with integration into the broader Cisco
    Security Cloud.
- aid: robust-intelligence:ai-firewall
  name: AI Firewall (Legacy)
  tags:
  - Runtime Protection
  - Guardrails
  - LLM Security
  - Legacy
  humanURL: https://www.robustintelligence.com/
  properties:
  - url: https://www.robustintelligence.com/
    type: Documentation
  description: The Robust Intelligence AI Firewall provided runtime guardrails for LLM and ML applications, screening prompts
    and responses for prompt injection, PII, toxicity, hallucination, and policy violations. Now integrated into Cisco
    AI Defense and rebranded under the Cisco product portfolio.
- aid: robust-intelligence:algorithmic-red-teaming
  name: Algorithmic AI Red Teaming (Legacy)
  tags:
  - Red Teaming
  - AI Security
  - Vulnerability Assessment
  - Legacy
  humanURL: https://www.robustintelligence.com/
  properties:
  - url: https://www.robustintelligence.com/
    type: Documentation
  description: Algorithmic AI Red Teaming was Robust Intelligence's automated adversarial testing product for ML and LLM
    models, generating attacks across data, model, and prompt layers and producing risk reports for security and ML teams.
    Capabilities now delivered through Cisco AI Defense.
common:
- type: Website
  url: https://www.robustintelligence.com/
- type: CiscoProduct
  url: https://www.cisco.com/site/us/en/products/security/ai-defense/index.html
- type: AcquisitionAnnouncement
  url: https://newsroom.cisco.com/c/r/newsroom/en/us/a/y2024/m08/cisco-announces-intent-to-acquire-robust-intelligence-to-deliver-comprehensive-ai-security.html
- type: Blog
  url: https://www.robustintelligence.com/blog
- type: LinkedIn
  url: https://www.linkedin.com/company/robust-intelligence/
- type: GitHubOrganization
  url: https://github.com/robustintelligence
- type: AcquiredBy
  url: https://www.cisco.com/
- type: Features
  data:
  - name: AI Firewall
    description: Runtime guardrails screening LLM prompts and responses for prompt injection, PII, toxicity, and policy
      violations.
  - name: Algorithmic Red Teaming
    description: Automated adversarial testing of ML and LLM models across data, model, and prompt attack surfaces.
  - name: Model Validation
    description: Pre-production validation of ML and LLM models against safety, security, and quality criteria.
  - name: Shadow AI Discovery
    description: Visibility into AI applications, models, and providers in use across the enterprise.
  - name: Cisco Security Cloud Integration
    description: Native integration with Cisco's broader security portfolio post-acquisition.
- type: UseCases
  data:
  - name: Enterprise AI Security
    description: Protect production LLM and ML applications against adversarial attacks and unsafe outputs.
  - name: AI Governance
    description: Inventory and govern AI usage across the enterprise to meet emerging regulatory requirements.
  - name: Pre-production Red Teaming
    description: Run automated adversarial assessments against AI models before deployment.
  - name: Runtime Guardrails
    description: Enforce policies on prompts and responses in production GenAI applications.
- type: Integrations
  data:
  - name: Cisco Security Cloud
    description: Native integration with Cisco's broader security platform.
  - name: OpenAI
    description: Guardrails and red teaming for OpenAI-based applications.
  - name: Anthropic
    description: Protection for Anthropic Claude-based applications.
  - name: Hugging Face
    description: Validation and red teaming of Hugging Face hosted models.
  - name: SageMaker
    description: Integration with AWS SageMaker for model validation workflows.
- type: AcquisitionNotice
  description: Robust Intelligence was acquired by Cisco in August 2024. The product is now delivered as Cisco AI Defense
    under the Cisco Security portfolio. Independent product names and developer-facing surfaces are being consolidated
    into Cisco's portals.
maintainers:
- FN: Kin Lane
  email: [email protected]