Alice Rebrands From ActiveFence to Become Core Safety Layer for AI and Social Platforms

New updates have been reported about Alice.

Alice, formerly known as ActiveFence, has formally repositioned itself as a dedicated safety and security layer for AI-powered and social platforms after nearly a decade operating largely behind the scenes. The company reports that its technology currently helps protect more than 3 billion users and underpins safety for 7 of the world’s 10 largest AI foundation models, working with major platforms and AI labs including Amazon, TikTok, Nvidia, Cohere, and Black Forest Labs. CEO and co-founder Noam Schwartz said the company’s core asset is its accumulated intelligence on how malicious behavior emerges and scales across online systems, arguing that effective AI security requires a deep understanding of human-driven abuse, not just technical controls. Alice defines its focus as “communicative tech” — systems that enable people to connect or interact with each other or with AI — and positions itself as an infrastructure layer enterprises can rely on as they rush to deploy AI capabilities without compromising safety or compliance.

At the center of Alice’s offering is Rabbit Hole, its proprietary adversarial intelligence engine built on years of real-world monitoring across cultures, languages, and threat types, from child exploitation and fraud to prompt-injection attacks and violent extremism. This engine feeds a suite of solutions across three pillars: Weaponized AI (detecting and mitigating scams, deepfakes, and manipulation), Interaction Security (real-time guardrails and red teaming to prevent model misuse and jailbreaks), and Model Integrity (ongoing evaluation and governance to keep model outputs aligned with policy, safety, and regulatory standards as they evolve). With eight years of live-environment data and continuous 24/7 threat monitoring, Alice aims to identify risks early, harden AI systems before exploitation, and prevent harm from reaching end users. Strategically, the rebranding signals a shift from being perceived primarily as a social media trust-and-safety vendor to a broader role as a foundational security layer for enterprises, user-generated content platforms, and frontier AI labs building generative AI applications and agents worldwide.