About
Alice, formerly known as ActiveFence, is a leading enterprise-grade security, safety, and trust platform built for the GenAI era. It offers the WonderSuite — a comprehensive set of tools that covers the full AI lifecycle: WonderBuild for pre-launch stress testing, WonderFence for dynamic runtime guardrails, and WonderCheck for ongoing automated red-teaming and drift detection in production environments. Alice is trusted by major foundation model labs and enterprises globally, safeguarding over 3 billion users across 120+ languages with more than 1 billion daily AI-human interactions monitored. Its competitive edge comes from the 'Rabbit Hole Data Moat' — a massive, continually updated repository of real-world adversarial examples that trains sharper classifiers and powers stronger red-teaming exercises. The platform addresses critical risks such as unintended GenAI behavior, prompt injection, jailbreaks, application-level exploits, governance and compliance gaps, and reputational exposure. Industries served include child-facing products, financial services, healthcare, insurance, and legal services. Alice enables organizations to align AI outputs with internal policies, regulatory requirements, and brand standards across every modality, empowering enterprises to innovate safely and confidently in the age of generative AI.
Key Features
- WonderBuild – Pre-Launch Stress Testing: Simulates adversarial attacks and edge cases before deployment to ensure AI models and agents are safe, resilient, and ready for production.
- WonderFence – Dynamic Runtime Guardrails: Enforces real-time policy compliance and safety rules on live AI applications and agents, blocking harmful or non-compliant interactions as they happen.
- WonderCheck – Continuous Red-Teaming: Automated, ongoing red-teaming in production to detect model drift, surface emerging risks, and prioritize remediation before issues escalate.
- Adversarial Intelligence Data Moat: Backed by billions of continually updated toxic and adversarial examples drawn from a decade of real-world research, powering superior classifier accuracy and coverage.
- Adaptive Policy Alignment: Customizable policy enforcement tuned to regulatory requirements and risk tolerance, covering every modality across multilingual and multi-platform environments.
Use Cases
- A foundation model lab uses WonderBuild to stress-test a new LLM for safety and policy compliance before public release, reducing the risk of harmful outputs at launch.
- An enterprise chatbot provider deploys WonderFence to enforce brand and regulatory policies in real time, automatically blocking jailbreak attempts and off-policy responses.
- A healthcare AI company uses WonderCheck to continuously monitor its production AI assistant for model drift and emerging safety risks, ensuring ongoing HIPAA alignment.
- A gaming platform with user-generated content leverages Alice's UGC Intelligence layer to detect and remove toxic or abusive content across 120+ languages at scale.
- A financial services firm uses Alice to ensure its AI-powered advisory tools remain compliant with financial regulations while protecting against adversarial exploitation.
Pros
- Proven at Massive Scale: Protects over 3 billion users and handles 1 billion+ daily AI-human interactions across 120+ languages, demonstrating enterprise-grade reliability.
- Full AI Lifecycle Coverage: WonderSuite covers pre-launch testing, live runtime protection, and continuous production monitoring — a true end-to-end safety solution.
- Deep Adversarial Data Advantage: A decade of real-world adversarial research and billions of threat examples give Alice's classifiers unmatched accuracy over generic competitors.
- Customizable Policy Enforcement: Flexible alignment to internal policies, brand standards, and regulatory frameworks across industries like healthcare, finance, and legal services.
Cons
- Enterprise-Only Pricing: Alice is positioned as an enterprise product with no self-serve or public pricing, making it inaccessible for smaller teams or startups.
- Requires Sales Engagement: Onboarding requires a demo and direct sales conversation, which can slow down evaluation for teams seeking quick integration.
- Overkill for Small-Scale AI Apps: The breadth and depth of WonderSuite may be more than necessary for small or low-risk AI deployments.
Frequently Asked Questions
What is Alice, formerly ActiveFence?
Alice is an enterprise AI trust and safety platform that provides pre-launch stress testing, real-time runtime guardrails, and continuous red-teaming to protect GenAI applications, agents, and UGC platforms from adversarial threats and policy violations.
What is WonderSuite?
WonderSuite is Alice's flagship security platform for enterprise AI, comprising three products: WonderBuild (pre-launch stress testing), WonderFence (dynamic runtime guardrails), and WonderCheck (ongoing automated red-teaming in production).
What industries does Alice serve?
Alice serves a wide range of regulated and high-risk industries including child-facing products, financial services, healthcare, insurance, and legal services.
How does Alice protect against prompt injection and jailbreaks?
WonderFence monitors live AI interactions in real time, detecting and blocking prompt injection attempts, jailbreaks, and model exploits before they can lead to data leakage or platform abuse.
How is Alice's data different from other AI safety tools?
Alice maintains a 'Rabbit Hole Data Moat' — a continually updated corpus of billions of real-world adversarial and toxic examples collected over a decade, which trains significantly more accurate and comprehensive safety classifiers than competitors relying on synthetic or limited datasets.
