The U.S. Cybersecurity and Infrastructure Security Agency (CISA) has recently released the AI Cybersecurity Collaboration Playbook, a guidance document designed to strengthen collaboration and information sharing within the Artificial Intelligence (AI) ecosystem to address cybersecurity threats effectively.

The playbook promotes collaboration among federal agencies, private industry, international partners, and other stakeholders to enhance awareness of AI cybersecurity risks and resilience. 

It guides CISA’s Joint Cyber Defense Collaborative (JCDC) partners on voluntary information sharing about AI-related cybersecurity incidents and vulnerabilities, explains protections and mechanisms for sharing, and details CISA’s response to shared information.

CISA clarifies that AI safety topics, such as risks to human life, health, property, or the environment, fall outside the playbook’s scope. However, the playbook is expected to evolve over time, adapting to emerging challenges through ongoing collaboration among government, industry, and international partners.

“This playbook does not create policies, impose requirements, mandate actions, or override existing legal or regulatory obligations. All actions taken under this playbook are voluntary,” CISA said in a blog post. 

In September last year, CISA issued guidance aimed at helping federal civilian agencies mitigate their cybersecurity risks.

The Federal Civilian Executive Branch Operational Cybersecurity Alignment plan encourages agencies to enhance their cyber capabilities by concentrating on asset management, vulnerability management, defensible architecture, supply chain resilience, and incident detection and response.