Coalition for Secure AI

What Coalition for Secure AI does

The Coalition for Secure AI (CoSAI) is an open consortium organized as an OASIS Open Project, founded in July 2024 at the Aspen Security Forum. It is not a commercial product vendor but rather a vendor-neutral standards body that develops security frameworks, taxonomies, open-source tools, and best-practice guidance for AI systems. The organization serves AI-deploying enterprises, security practitioners, developers, government/regulatory bodies, and academic researchers, organizing its work across four workstreams: Software Supply Chain Security for AI Systems, Preparing Defenders for a Changing Cybersecurity Landscape, AI Security Risk Governance, and Secure Design Patterns for Agentic Systems.

Its core deliverables include the AI Shared Responsibility Framework (a five-layer accountability model), the Agentic Identity and Access Management Framework, the Model Context Protocol (MCP) Security taxonomy (12 threat categories, ~40 individual threats), the AI Incident Response Framework, the Open Model Signing Specification (Sigstore-based signatures for ML model artifacts), Project CodeGuard (an AI model-agnostic security coding framework donated by Cisco), the CoSAI Risk Map, and frameworks addressing AI supply chain risks and defender preparation. All technical outputs are freely distributed via GitHub and the CoSAI website.

CoSAI generates revenue through tiered annual membership dues rather than product sales. Premier Sponsors pay $70,000 annually for governing board representation, while General Sponsors pay scaled fees from $31,500 (enterprises with 2,000+ employees) down to $1,300 (nonprofits and academic institutions in developing economies). Technical participation is free and open to all developers regardless of sponsorship status. The organization has attracted 40+ partner organizations including Google, Microsoft, NVIDIA, IBM, Meta, Amazon, Anthropic, OpenAI, Cisco, Intel, PayPal, Trend Micro, Zscaler, and Humana, and is governed by a Project Governing Board co-chaired by representatives from Cisco (Omar Santos) and Google (David LaBianca), with technical direction set by a Technical Steering Committee.

Name

Coalition for Secure AI

Legal name

Coalition for Secure AI

Website

https://coalitionforsecureai.org

Company type

Private

Founded year

2024

Operating status

Operating

Headcount range

1–10 employees

Short description

CoSAI is an OASIS Open-hosted consortium that develops vendor-neutral security frameworks, taxonomies, and open-source tools for AI systems. It serves AI-deploying enterprises, security practitioners, and governments through tiered annual membership dues and free public technical output.

Ownership category

akta.pro rank

Product category

AI Security Frameworks and Standards

NAICS

Computer Systems Design and Related Services (54151)

SIC

Services-Computer Integrated Systems Design (7373)

akta.pro primary industry

AI Governance, Risk & Compliance (GRC) Platforms (HDAAAMAA)

akta.pro secondary industries

Secure Model Deployment & Runtime Protection (sandboxing, isolation) (HDAAAKAH), Identity, Access & Secrets Management for AI Systems (IAM for agents/models) (HDAAAKAJ), Responsible AI, Security & Privacy Platforms (Safety, Guardrails, PII) (HDAEANAG)

Keywords

• AI security frameworks
• Open consortium standards
• AI risk governance
• Agentic AI security
• Model signing specifications

GTM type

B2B

Offering type

Software

Cost components

Personnel, Operations, Marketing or Sales, Technology or R&D, Others

Revenue model

1. Membership Dues: CoSAI generates revenue through annual membership dues structured by organization size and type. Premier Sponsors pay $70,000 annually with full governance benefits including seat on Governing Board. General Sponsors pay scaled fees from $31,500 (2,000+ employees) down to $1,300 (nonprofit/academic from developing economies). Membership provides governance representation, visibility benefits, and OASIS Open membership perks.

Pricing tiers

Go-to-market motion1 record

Distribution channels2 records

Marketing channels5 records

Core offering

Coalition for Secure AI (CoSAI) is an open consortium operating under OASIS Open that develops standardized security frameworks, taxonomies, specifications, and open-source tools for AI systems. Its deliverables include security frameworks (AI Shared Responsibility, Agentic IAM, MCP Security, AI Incident Response), technical specifications (Open Model Signing), and open-source tools (Project CodeGuard, CoSAI Risk Map), all freely distributed through GitHub and the CoSAI website to enable secure AI development and deployment across the industry.

Product overview

Coalition for Secure AI (CoSAI) is an open ecosystem organized as an OASIS Open Project that develops frameworks, specifications, and open-source tools rather than a commercial product. The portfolio includes security frameworks such as the AI Shared Responsibility Framework (five-layer accountability model), Agentic Identity and Access Management Framework, Model Context Protocol Security taxonomy, and AI Incident Response Framework. Technical specifications include the Open Model Signing Specification for ML artifact signatures. Open-source tools include Project CodeGuard (security coding framework) and the CoSAI Risk Map (taxonomy tool). Four workstreams organize the technical work: Software Supply Chain Security for AI Systems, Preparing Defenders for a Changing Security Landscape, AI Security Risk Governance, and Secure Design Patterns for Agentic Systems. All outputs are freely available through GitHub and the CoSAI website.

Differentiator

Problem solved

Functional benefit

Products and services

• AI Shared Responsibility Framework
• Agentic Identity and Access Management Framework
• Model Context Protocol (MCP) Security Framework
• The Future of Agentic Security: From Chatbots to Autonomous Swarms
• AI Incident Response Framework
• Open Model Signing Specification
• Project CodeGuard
• CoSAI Risk Map
• Risks and Controls for the AI Supply Chain
• Preparing Defenders of AI Systems

Quantifiable outcome

• 12 threat categories identified for MCP security
• +2 more outcomes

Segments4 records

Ideal customer profiles3 records

Technology focussed Yes

API detail

Has API

No

API docs

API detail

Core technology

AI maturity

App detail

AI capability5 records

Feature7 records

Partnerships

20 partnerships are on record, tiered flagship and core.

• HumanaflagshipStrategic or Co-development Partner · 1 April 2026Humana joined as Premier Sponsor bringing healthcare sector perspective to AI security. James Norberg serves as VP of Technical Product Security Services and Information Security Officer for Humana's Retail business, contributing expertise across healthcare, financial services, and critical infrastructure.
• MetaflagshipStrategic or Co-development Partner · 3 February 2026Meta joined as Premier Sponsor in February 2026 to advance industry-wide AI security standards. Scott Bratsman serves as Senior Director of Product Management for Cybersecurity at Meta. Contributes expertise in AI development and deployment to CoSAI's collaborative ecosystem.
• GoogleflagshipStrategic or Co-development Partner · 1 July 2024Google is a Premier Sponsor and founding member of CoSAI, represented on the Project Governing Board and Executive Steering Committee. David LaBianca serves as PGB co-chair. Google contributes to AI security research through its security engineering team including Heather Adkins, VP and Cybersecurity Resilience Officer.
• MicrosoftflagshipStrategic or Co-development Partner · 1 July 2024Microsoft is a Premier Sponsor and founding member of CoSAI. Nicholas Butts serves as Director of Cybersecurity and AI Security Policy. Yonatan Zunger serves as CVP of AI Safety & Security. Contributes to workstreams on AI security standards and secure development practices.
• NVIDIAflagshipStrategic or Co-development Partner · 1 July 2024NVIDIA is a Premier Sponsor and founding member represented by Daniel Rohrer, VP of Software Product Security, Architecture and Research. Committed to building a community dedicated to making secure and trustworthy AI accessible to all.
• IBMflagshipStrategic or Co-development Partner · 1 July 2024IBM is a Premier Sponsor and founding member. J.R. Rao serves as IBM Fellow and CTO of Security Research leading IBM's involvement. The Security Research team comprises over 200 researchers. IBM contributes to all workstreams including WS3 on AI Security Risk Governance.
• EYflagshipStrategic or Co-development Partner · 1 July 2024EY is a Premier Sponsor committed to fostering innovation while enhancing security and integrity of AI technologies. Yang Shim serves as EY Americas Technology Consulting Leader and Kapish Vanvaria as EY Americas Risk Leader. Sarah Liang serves as EY Global Responsible AI Leader.
• PayPalflagshipStrategic or Co-development Partner · 1 July 2024PayPal is a Premier Sponsor working to shape industry guidelines and standards for secure AI development. Shaun Khalfan serves as Chief Information Security Officer. PayPal contributes expertise from powering approximately a quarter of the world's e-commerce transactions annually.
• Trend MicroflagshipStrategic or Co-development Partner · 1 July 2024Trend Micro is a Premier Sponsor committed to leading the charge in securing AI deployment. Eva Chen serves as CEO. David Girard serves as Senior Director of Product Management for AI Security and AI Alliances. Contributes to secure design patterns and agentic systems work.
• ZscalerflagshipStrategic or Co-development Partner · 1 July 2024Zscaler is a Premier Sponsor collaborating to establish best practices ensuring AI technologies are innovative and trustworthy. Deepen Desai serves as Chief Security Officer. Contributes advanced AI-driven security solutions expertise.
• SnykflagshipStrategic or Co-development Partner · 1 July 2024Snyk is a Premier Sponsor contributing to AI security standards development. Participates in collaborative efforts to create open frameworks for building secure AI systems.
• OpenAIcoreStrategic or Co-development Partner · 1 July 2024OpenAI is a member organization contributing to CoSAI's work on AI security. Nick Hamilton serves as Head of Governance, Risk, and Compliance at OpenAI. Participates in industry collaboration on AI safety standards.
• AmazoncoreStrategic or Co-development Partner · 1 July 2024Amazon is a General Sponsor with active participation in CoSAI workstreams. Matt Saner serves as security leader at AWS focused on helping enterprises transform security challenges. Contributes to secure AI adoption guidance.
• AnthropiccoreStrategic or Co-development Partner · 1 July 2024Anthropic is a General Sponsor actively contributing to CoSAI's technical work. Jason Clinton serves as CISO and presented at RSAC 2026 on MCP security. Akila Srinivasan from Anthropic co-presented sessions on securing enterprise AI.
• CiscocoreStrategic or Co-development Partner · 1 July 2024Cisco donated Project CodeGuard to CoSAI in February 2026, an AI model-agnostic security coding framework. Omar Santos serves as Distinguished Engineer and PGB co-chair. Cisco contributes significantly to WS4 on secure design patterns for agentic systems.
• IntelcoreStrategic or Co-development Partner · 1 July 2024Intel is a founding member and General Sponsor. Dhinesh Manoharan serves as VP & GM of Security for AI & Security Research. Intel contributes to advancing secure AI practices and provides accessible guidance, resources and tools for secure AI systems.
• Red HatcoreStrategic or Co-development Partner · 1 July 2024Red Hat is a General Sponsor contributing to AI security standards. Vincent Danen serves as Vice President of Product Security and is a governing board member of OpenSSF. Contributes open source security expertise.
• WizcoreStrategic or Co-development Partner · 1 July 2024Wiz is a Premier Sponsor contributing AI and threat research expertise. Alon Schindel serves as VP of AI & Threat Research. Contributes to detection, cloud risks, threat intelligence and AI security research.
• HackerOnecoreStrategic or Co-development Partner · 1 July 2024HackerOne is a General Sponsor contributing security expertise. Nidhi Aggarwal serves as Chief Product Officer leading the platform that combines human expertise with AI to help enterprises secure software at scale.
• OASIS OpenflagshipStrategic or Co-development Partner · 1 July 2024CoSAI operates under OASIS Open, the international standards and open source consortium. OASIS provides governance, IP management, collaboration tools, and administrative services. All CoSAI work falls under OASIS project governance framework.

Scale indicators6 records

Direct peers

• Cloud Security Alliance (CSA): Cloud Security Alliance operates the AI Safety Initiative and spun out the CSAI Foundation, producing AI security research, white papers (30+), and certifications like TAISE. Directly comparable to CoSAI as a vendor-neutral consortium developing AI security frameworks and certifications.
• Linux Foundation AI & Data: Linux Foundation AI & Data hosts open source AI projects and standards initiatives. Comparable as a neutral, vendor-funded consortium stewarding AI infrastructure and governance projects with tiered membership and cross-industry sponsorship.
• Partnership on AI: Partnership on AI is a multi-stakeholder nonprofit developing AI best practices across safety, fairness, and governance. Comparable as a cross-industry coalition producing AI guidance, though more focused on responsible AI broadly than security specifically.
• OWASP Foundation: OWASP produces the widely adopted OWASP Top 10 for LLM Applications and AI Security & Privacy Guide. Both CoSAI and OWASP are open communities developing AI security standards, frameworks, and best-practice documentation for practitioners.
• MITRE Corporation: MITRE maintains ATLAS (Adversarial Threat Landscape for AI Systems), a knowledge base of adversary tactics against AI systems with case studies. Directly comparable threat taxonomy work that informs defensive AI security programs.
• Frontier Model Forum: Frontier Model Forum is an industry consortium of leading AI labs (including Anthropic, Google, Microsoft, OpenAI — many CoSAI sponsors) focused on safe and responsible frontier model development. Comparable mission in AI safety standards with overlapping membership.
• OpenSSF (Linux Foundation): OpenSSF brings together industry stakeholders under Linux Foundation governance to improve open source software security, including AI supply chain work. CoSAI's WS1 on Software Supply Chain Security for AI overlaps directly with OpenSSF's SLSA and supply chain initiatives.

Emerging players

• Adversa AI: Adversa AI is a General Sponsor of CoSAI and operates as an emerging AI security research firm focused on adversarial ML and AI red teaming. Comparable niche as an emerging player in AI security research contributing to and consuming AI security standards.

Broad incumbents

• ISO/IEC SC 42 (Artificial Intelligence): ISO/IEC SC 42 develops international AI standards covering governance, risk management, and security. CoSAI's frameworks could be fast-tracked through ISO's PAS process via OASIS, making SC 42 a relevant incumbent peer in the AI standards landscape.
• NIST AI Risk Management Framework: NIST's AI RMF and AI Safety Institute consortium provide government-backed AI risk governance frameworks. Comparable in mission (standardizing AI risk management) but with regulatory authority and broader scope than CoSAI's consortium-based approach.

Market position

Strengths1 record

Weaknesses4 records

Key risks5 records

Key highlights6 records

Customer concentration

• GitHubgithub.com/cosai-oasis
• LinkedInlinkedin.com/company/coalition-for-secure-ai

Revenue estimate

Valuation estimate

Number of profiles

Profiles2 records

Funding overview

Funding rounds

Investors

Funding detail is available on the Subscription and Enterprise plan.Contact sales →

M&A

Investments

M&A and investment is available on the Subscription and Enterprise plan.Contact sales →

Frequently asked questions about Coalition for Secure AI