As AI and Generative AI workloads grow, ensuring the safety and security of AI workloads is more important than ever. Users need AI models to align with established protocols in safety and security, but this shouldn’t come at the cost of speed, performance, or increased costs. To tackle this challenge, Rivos and Virtue AI are collaborating to accelerate Virtue AI’s safety and security framework, VirtueGuard, by leveraging Rivos’ high-performance hardware and software solutions.
Rivos: Powering AI Workloads
Rivos develops hardware and software solutions, designed specifically to meet the needs of AI workloads. Their architecture is flexible and power-efficient, built to support the demanding needs of AI models as they evolve. From large language models to complex data analytics, Rivos’ hardware and software solutions scale effortlessly to meet the requirements of both current and future AI systems.
Virtue AI: Securing AI with VirtueGuard
VirtueGuard is Virtue AI’s guardrail model, offering policy-compliant, multimodal, multilingual moderation that is both efficient and highly effective. It ensures AI applications maintain high safety and security standards that are aligned with established protocols.
VirtueGuard runs seamlessly on Rivos
Rivos has simplified the process of hardware adoption for customers and partners by using a workload-driven hardware design philosophy. VirtueGuard ran seamlessly on Rivos’ architecture using Rivos’ intuitive toolchain without any redesign—just a simple recompile.
Meeting the Evolving Needs of AI
In today’s rapidly evolving AI landscape, ensuring AI safety and security is essential. Real-time moderation plays a critical role in maintaining this level of safety. For example, chatbots must respond quickly while safeguarding against harmful requests and security risks. As these AI models grow in size and complexity, the processing requirements on underlying data centers increase significantly. Striking a balance between handling these new, larger models and maintaining the fast inference speeds without compromising confidence in the final outcomes is essential.
Rivos and Virtue AI provide a proactive framework that maintains high performance without sacrificing security. Equipped with advanced contextual awareness and dynamic risk assessment capabilities, VirtueGuard not only prevents harmful or inappropriate content but also adapts to emerging threats, offering a proactive defense against evolving challenges. Rivos’ architecture enables VirtueGuard to run seamlessly with just a recompile. Together, they ensure that AI systems can evolve while still maintaining user trust and security.
A Commitment to Trustworthy AI
“As the AI landscape continues to diversify and expand, building confidence and trust in the underlying systems becomes more critical than ever,” says Puneet Kumar, CEO and co-founder of Rivos Inc., “However, this must be achieved efficiently and in near real-time, without placing an excessive burden on the data processing. By collaborating, Virtue AI and Rivos can deliver a secure, processing-efficient solution that is both flexible and capable of ensuring safety across a wide range of models.”
“We built VirtueGuard based on our deep understanding about vulnerabilities of AI systems,” said Bo Li, CEO of Virtue AI Inc. “It is straightforward to deploy VirturGuard on the Rivos architecture. Our collaboration with Rivos represents a strategic effort to build scalable, practical solutions that provide safety protections on efficient and performant hardware. Together, we aim to create AI systems that can meet the rigorous demands of modern applications while maintaining safety, reliability, and trustworthiness.”
About Rivos
Rivos has assembled a world-class team of silicon, software and platform designers implementing the long-term vision of building industry-leading power efficient, high performance, secure server solutions, based on RISC-V, using workload-defined hardware. Rivos supports the intense requirements of the large language models and data analytics through a full solution of optimized chips; combining RISC-V CPUs, a Data Parallel Accelerator, and a reference multi-chip OCP modular server, along with a full firmware-to-application open software stack. Rivos is hiring engineering talent across multiple disciplines.
About Virtue AI
Virtue AI is an AI Safety and Security startup, founded by leading researchers from Stanford, UC Berkeley and UIUC. Virtue AI’s goal is to help enterprises spend more time building products, and less time worrying about safety, security, and compliance. Virtue AI has packaged state-of-the-art research into solutions that enterprises can deploy out-of-the-box for their bespoke needs in red-teaming or developing guardrails. Request a demo here. Virtue AI is hiring.
