On Tuesday, AI safety and security company Alice announced a significant milestone by securing $140 million in a recent funding round. This capital aims to bolster their efforts in protecting AI systems from new vulnerabilities and adversarial threats, marking a crucial step in the evolution of AI defenses.
Strategic Use of New Funding
The newly acquired funds raise Alice’s total capital to $280 million, which will be directed towards enhancing their AI platform. Key focuses include expanding their team dedicated to proprietary intelligence datasets and scaling their market strategies. These strategic moves are expected to fortify the company’s position in the AI trust and safety sector.
This funding round was spearheaded by Apax Digital Funds, with substantial contributions from notable investors such as MoreTech, Phoenix Financial, and several others. Their involvement underscores the growing interest and confidence in Alice’s mission to secure AI technologies.
Innovative Security Solutions
Alice collaborates closely with AI developers to rigorously test foundational models before they are publicly released. The company’s research initiatives involve simulating malicious inputs and complex tasks to identify and rectify erratic behaviors in AI models. This process is critical in safeguarding against prompt injection and jailbreak attempts.
Once models are deployed, Alice provides ongoing red-teaming and dynamic runtime guardrails, allowing businesses to align AI operations with specific protocols. The platform supports custom guideline configurations, breach simulations, and real-time system traffic monitoring.
Research and Expertise
Central to Alice’s capabilities is its proprietary Rabbit Hole database, developed over nearly ten years to track digital fraud, extremist activities, and manipulation efforts. This extensive archive enables Alice to effectively detect and counteract threats against generative AI systems.
A dedicated research facility is home to over 150 specialists who collaborate with leading model developers. Their work focuses on identifying potential vulnerabilities, analyzing manipulation techniques, and proactively addressing security weaknesses in machine learning architectures.
According to Noam Schwartz, CEO and co-founder of Alice, the company has spent nearly a decade understanding how technology can be exploited on a massive scale. With this funding, Alice aims to close the gap between the rapid democratization of AI capabilities and the defenses needed to protect them.
Previously known as ActiveFence, Alice operates from its headquarters in New York and Tel Aviv, continuing to drive innovation in AI security measures.
