Intel Joins Coalition for Secure AI (CoSAI) to Lead in Secure AI Development

A Unified Approach to Secure AI

Intel proudly announces its founding membership in the Coalition for Secure AI (CoSAI). This initiative aims to establish a collaborative ecosystem for the secure development and deployment of artificial intelligence.

As AI rapidly transforms industries, developers and users alike must navigate a maze of inconsistent standards and fragmented practices. To tackle this, Intel is working alongside other leaders to ensure AI systems are built with security at their core.

Why AI Security Demands Collaboration

Securing AI technology requires more than just patches and policies. In fact, it demands unified standards, transparent governance, and continuous innovation. Intel believes collaboration is critical to address future security challenges especially as AI becomes increasingly complex.

At Intel, we’ve always led in shaping secure, innovative technologies. From experience, we know our customers and partners face growing pressures to adopt AI responsibly. To meet these demands, shared frameworks and security-first development practices are essential.

Pat Gelsinger’s Roadmap for Scalable, Secure AI

At Intel Vision 2024, CEO Pat Gelsinger presented a roadmap for open, scalable AI systems. His plan outlines the tools, hardware, software, and methodologies required for building trustworthy AI solutions.

In line with this vision, Intel joined Google, IBM, and other major firms as a founding member of CoSAI. Hosted by OASIS Open, CoSAI focuses on creating resources that help developers design AI systems that are secure by design.

CoSAI’s Mission and Key Work Streams

CoSAI brings together global experts from industry, academia, and government. Their shared goal is to create best practices, frameworks, and tools for secure AI development.

The coalition will focus on three strategic areas:

• Software supply chain security: Improving composition and provenance tracking to harden AI applications.

• Preparing defenders: Addressing security integration in AI and traditional systems for better resilience.

• AI governance and risk: Developing actionable best practices and robust assessment frameworks.

These collaborative efforts ensure that AI security evolves in step with the technology itself.

Intel’s Continued Commitment to Responsible AI

Intel is dedicated to advancing secure AI development through proactive leadership and industry cooperation. As part of CoSAI, we will continue sharing our security assurance expertise to strengthen AI systems worldwide.

Moreover, our work with CoSAI aligns with other ongoing initiatives. For example, the Open Platform for Enterprise AI (OPEA) a Linux Foundation AI & Data Sandbox Project supports safe, scalable GenAI implementations. Intel is also a founding member of OPEA, with efforts starting around retrieval-augmented generation (RAG).

Building a Future of Trustworthy AI

The AI landscape is growing fast. To keep pace, technology vendors must commit to open, secure, and flexible solutions. That’s why Intel will continue delivering reliable, security-enhanced products that enable responsible AI adoption.

In conclusion, Intel’s role in CoSAI reflects its broader vision: to ensure AI is not just powerful but also trustworthy, transparent, and secure.

secure AI development