Skip to main content
Rase AI Intelligence Newsletter

Safety & governance Product

Google Cloud previews unified access policies for humans and AI agents

Google IAM records Unified Access Policies entering preview on November 2. A later technical article explains unified allow and deny rules and support for human, workload and agent identities. [1][2]

Release status
Preview availability recorded November 2; technical explainer published November 20. [1][2]

What it does

The November 20 technical article describes reusable policies across organization, folder and project scopes, concurrent evaluation with existing IAM, and management through the console, command line, Terraform and REST interfaces. Those details describe the current preview, rather than establishing its first availability that day. [1][2]

Why it matters

Security teams managing growing numbers of agents can express shared access boundaries without duplicating every rule across resources. [1][2]

Who it’s for

Cloud platform, IAM and agent administrators. [1][2]

Key dates

  1. Google IAM documents Unified Access Policies preview availability [1][2]

Where it fits

Evaluate preview limits and how new policies interact with existing allow and deny rules before adopting them. [1][2]

Limits and trade-offs

This is a preview, not a generally available security guarantee. Policy limits and interaction with existing IAM require review. The November 20 article is an explainer of a preview already recorded on November 2, rather than a second launch. Its stated lack of additional preview cost does not make all related cloud usage free. [1][2]

Sources and verification

Catalog evidence checked How we verify records

  1. Primary source How to simplify cloud access with Unified Access Policies, now in preview
  2. Primary source Google Cloud IAM release notes: November 2 Unified Access Policies preview

Related profiles

Safety & governance

Zscaler frontier-model security collaborations

Zscaler

The issuer says Claude Mythos 5.1 is integrated into Endpoint AI Security for its Exploit Paths capability, with a limited preview planned. It also describes a Mistral integration through its European sovereign Zero Trust Exchange.

High confidenceEvidence checked

Safety & governance

EU–UK Digital Partnership plan

European Union and United Kingdom

Paragraph54 of the joint statement identifies AI, including safety and security, among possible cooperation areas alongside connectivity, quantum technology, digital credentials and online safety.

High confidenceEvidence checked

Safety & governance

GAO AI and cybersecurity technology assessment

U.S. Government Accountability Office

GAO reviewed research and agency material, interviewed stakeholders and convened experts. Its options address guidance, research, lifecycle governance and workforce capability. Examples include broader testing environments, output validation…

High confidenceEvidence checked