July 19, 2024 at 10:12AM
Google has launched the Coalition for Secure AI (CoSAI), in partnership with industry players like Amazon, IBM, and Microsoft, to address cybersecurity risks in artificial intelligence. CoSAI aims to establish common security standards, provide guidance on evaluating software supply chains, and develop frameworks for identifying and mitigating AI security impacts. The initiative also looks to collaborate with other organizations to promote responsible AI.
From the provided meeting notes, it is clear that Google has launched the Coalition for Secure AI (CoSAI) as an initiative to address cybersecurity risks associated with artificial intelligence (AI). CoSAI is housed under OASIS Open and includes founding members such as Amazon, Anthropic, Chainguard, Cisco, Cohere, GenLab, IBM, Intel, Microsoft, NVIDIA, OpenAI, PayPal, and Wiz.
The coalition aims to advance comprehensive security measures for addressing the unique risks associated with AI in real time and over the horizon. CoSAI will focus on three main areas: software supply chain, framework for defenders to address security impact of AI, and guidance development for practitioners.
Furthermore, CoSAI will collaborate with industry and academia to establish common security standards and best practices in AI security. It will also work with organizations such as the Frontier Model Forum, ML Commons, Open Source Security Foundation, and Partnership on AI to advance responsible AI.
Overall, CoSAI seeks to provide a framework for AI security that responsibly captures the opportunity in front of us, with more updates expected in the coming months.