<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"><channel><title>AI Agent Access Control</title><link>https://aiagentaccesscontrol.com/</link><description>AI Agent Access Control maps the identity, default permissions and boundaries of AI agent platforms in AWS, Azure and Google Cloud, with a 30-check checklist.</description><lastBuildDate>Tue, 10 Nov 2026 00:00:00 GMT</lastBuildDate>
<item><title>The identity an AI agent runs as in AWS, Google Cloud and Azure</title><link>https://aiagentaccesscontrol.com/guides/the-identity-an-agent-runs-as/</link><guid>https://aiagentaccesscontrol.com/guides/the-identity-an-agent-runs-as/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>An agent has no access of its own. Every call it makes is judged against the identity it runs as: an IAM role on AWS, a service account or agent identity o</description></item>
<item><title>Why an AI agent's permissions are not its user's permissions</title><link>https://aiagentaccesscontrol.com/guides/why-an-agents-permissions-are-not-its-users/</link><guid>https://aiagentaccesscontrol.com/guides/why-an-agents-permissions-are-not-its-users/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>A user who talks to an agent does not lend it their access. The agent calls tools with its own identity unless the platform exchanges tokens to act for the</description></item>
<item><title>Network reachability for AI agents on AWS, Google Cloud and Azure</title><link>https://aiagentaccesscontrol.com/guides/network-reachability-for-agents/</link><guid>https://aiagentaccesscontrol.com/guides/network-reachability-for-agents/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>What an agent can reach over the network depends on where it runs and what you attach. This guide covers private endpoints, VPC attachment and egress filte</description></item>
<item><title>Logging what an AI agent did: audit, data and model logs</title><link>https://aiagentaccesscontrol.com/guides/logging-what-an-agent-did/</link><guid>https://aiagentaccesscontrol.com/guides/logging-what-an-agent-did/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>After an incident the question is which agent did what, to which data. This guide covers the three log layers on AWS, Google Cloud and Azure (control plane</description></item>
<item><title>AI agent permissions: how to cut an agent's access</title><link>https://aiagentaccesscontrol.com/guides/least-privilege-for-ai-agents/</link><guid>https://aiagentaccesscontrol.com/guides/least-privilege-for-ai-agents/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AI agent permissions start too wide when a starter role or broad managed policy gets the agent working fast. This guide shows how to write the policy from </description></item>
<item><title>Amazon Bedrock agent permissions: roles, trust and what the console creates</title><link>https://aiagentaccesscontrol.com/guides/bedrock-agent-permissions/</link><guid>https://aiagentaccesscontrol.com/guides/bedrock-agent-permissions/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Bedrock agents run as IAM roles that the service assumes. This guide covers Agents Classic and AgentCore: the trust policy and its condition keys, the docu</description></item>
<item><title>Vertex AI agent service account: Agent Runtime identities in Google Cloud</title><link>https://aiagentaccesscontrol.com/guides/vertex-ai-agent-service-account/</link><guid>https://aiagentaccesscontrol.com/guides/vertex-ai-agent-service-account/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Agent Runtime, formerly Vertex AI Agent Engine, lets an agent run as a per-agent identity, a Google-managed service agent or a custom service account. This</description></item>
<item><title>What can an AI agent access in AWS: the four layers to read</title><link>https://aiagentaccesscontrol.com/guides/what-can-an-ai-agent-access-in-aws/</link><guid>https://aiagentaccesscontrol.com/guides/what-can-an-ai-agent-access-in-aws/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>To answer what an agent can reach in an AWS account, read four layers: the role it assumes, the tools and Lambda functions behind it, the resource-based po</description></item>
<item><title>Approving an AI agent for production: running the review meeting</title><link>https://aiagentaccesscontrol.com/guides/approving-an-agent-for-production/</link><guid>https://aiagentaccesscontrol.com/guides/approving-an-agent-for-production/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>A production review for an AI agent works when it follows a fixed order and ends with recorded decisions. This guide gives the pre-read, the agenda built o</description></item>
<item><title>AI agent permissions in Azure: Foundry Agent Service, identities and RBAC</title><link>https://aiagentaccesscontrol.com/guides/ai-agent-permissions-in-azure/</link><guid>https://aiagentaccesscontrol.com/guides/ai-agent-permissions-in-azure/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Microsoft Foundry Agent Service gives each agent a Microsoft Entra agent identity that starts without roles. This guide covers agent identities, project an</description></item>
<item><title>AI agent authorization in AWS, Azure and Google Cloud</title><link>https://aiagentaccesscontrol.com/ai-agent-authorization/</link><guid>https://aiagentaccesscontrol.com/ai-agent-authorization/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AI agent authorization in a cloud account is one decision made on up to three inputs: the identity the agent runs as, the policies attached to the resource</description></item>
<item><title>AI agent guardrails: SCPs, Azure Policy and org policy</title><link>https://aiagentaccesscontrol.com/ai-agent-guardrails/</link><guid>https://aiagentaccesscontrol.com/ai-agent-guardrails/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AI agent guardrails for cloud access are the limits set above the agent's own role: AWS SCPs, RCPs and permissions boundaries, Azure Policy and deny assign</description></item>
<item><title>AI access control in cloud accounts: roles, policy, network</title><link>https://aiagentaccesscontrol.com/ai-access-control/</link><guid>https://aiagentaccesscontrol.com/ai-access-control/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AI access control for AI workloads in a cloud account has four layers: the identity the workload runs as, the policies that grant it access, the ceilings a</description></item>
<item><title>Agent authorization patterns: five ways to scope an agent</title><link>https://aiagentaccesscontrol.com/agent-authorization-patterns/</link><guid>https://aiagentaccesscontrol.com/agent-authorization-patterns/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Five authorization patterns scope an AI agent in a cloud account: a dedicated role per agent, a role per tool, session policies that scope down credentials</description></item>
<item><title>Amazon Q Business default permissions: roles it creates</title><link>https://aiagentaccesscontrol.com/default-permissions/amazon-q-business/</link><guid>https://aiagentaccesscontrol.com/default-permissions/amazon-q-business/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Amazon Q Business runs as separate IAM service roles for the application, web experience, data sources and plugins, which the console creates if you let it</description></item>
<item><title>Copilot Studio default permissions: tool authentication</title><link>https://aiagentaccesscontrol.com/default-permissions/copilot-studio/</link><guid>https://aiagentaccesscontrol.com/default-permissions/copilot-studio/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>In Microsoft Copilot Studio an agent's reach depends on how each tool authenticates: agent author authentication, which Microsoft recommends for low-risk o</description></item>
<item><title>Azure OpenAI Assistants default permissions and access</title><link>https://aiagentaccesscontrol.com/default-permissions/azure-openai-assistants/</link><guid>https://aiagentaccesscontrol.com/default-permissions/azure-openai-assistants/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Azure OpenAI Assistants (Preview) is documented by Microsoft as retired in favor of Foundry Agent Service. The quickstart authenticates with an API key; En</description></item>
<item><title>Conversational Agents (Dialogflow CX) default permissions</title><link>https://aiagentaccesscontrol.com/default-permissions/conversational-agents/</link><guid>https://aiagentaccesscontrol.com/default-permissions/conversational-agents/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Conversational Agents (Dialogflow CX) calls tools and webhooks as the Dialogflow CX Service Agent, or as a service account you name. Google documents that </description></item>
<item><title>AWS Nitro Sandboxes give pods, tasks and agents their own hardware-isolated VM</title><link>https://aiagentaccesscontrol.com/news/nitro-sandboxes-eks-ecs-ec2-agents/</link><guid>https://aiagentaccesscontrol.com/news/nitro-sandboxes-eks-ecs-ec2-agents/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AWS announced AWS Nitro Sandboxes on 9 November 2026: lightweight VMs managed by the Nitro Hypervisor for an EKS pod, ECS task or VM, including AI agents r</description></item>
<item><title>Amazon Quick adds smart approvals for actions that agents take</title><link>https://aiagentaccesscontrol.com/news/quick-smart-approvals-agentic-actions/</link><guid>https://aiagentaccesscontrol.com/news/quick-smart-approvals-agentic-actions/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AWS announced on 28 October 2026 that Amazon Quick supports smart approvals. A separate classifier model reviews each eligible agent action, approves routi</description></item>
<item><title>WorkSpaces for AI agents adds OAuth sessions inside existing network boundaries</title><link>https://aiagentaccesscontrol.com/news/workspaces-for-ai-agents-oauth-sessions/</link><guid>https://aiagentaccesscontrol.com/news/workspaces-for-ai-agents-oauth-sessions/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AWS said on 20 October 2026 that WorkSpaces for AI agents now handles authentication and desktop connections in its managed MCP service. Sessions use exist</description></item>
<item><title>Act Security and HUB International build a no-credentials coding agent environment</title><link>https://aiagentaccesscontrol.com/news/act-hub-international-agentcore-coding-agents/</link><guid>https://aiagentaccesscontrol.com/news/act-hub-international-agentcore-coding-agents/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Act Security published a blog post on 19 October 2026 about HUB International moving Claude Code into cloud-hosted sandboxes on Amazon Bedrock AgentCore in</description></item>
<item><title>Act Security argues AI governance should limit what agents can reach</title><link>https://aiagentaccesscontrol.com/news/act-real-ai-risk-what-agents-can-reach/</link><guid>https://aiagentaccesscontrol.com/news/act-real-ai-risk-what-agents-can-reach/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Act Security published a news piece on 18 October 2026 titled The Real AI Risk Isn't How Agents Think. It's What They Can Reach. Act argues that agent reas</description></item>
<item><title>Amazon Bedrock Managed Agents powered by OpenAI enters preview</title><link>https://aiagentaccesscontrol.com/news/bedrock-managed-agents-openai-preview/</link><guid>https://aiagentaccesscontrol.com/news/bedrock-managed-agents-openai-preview/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AWS announced on 29 September 2026 a preview of Amazon Bedrock Managed Agents, built jointly with OpenAI. AWS says each agent runs with its own IAM role, s</description></item>
<item><title>Security Command Center flags reasoning engines that can modify IAM policies</title><link>https://aiagentaccesscontrol.com/news/scc-risk-engine-reasoning-engines-modify-iam/</link><guid>https://aiagentaccesscontrol.com/news/scc-risk-engine-reasoning-engines-modify-iam/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Google Cloud said on 28 September 2026 that the Security Command Center Risk Engine now detects and reports reasoning engines that can modify IAM policies </description></item>
<item><title>IAM outbound identity federation supports VPC endpoints for OIDC discovery</title><link>https://aiagentaccesscontrol.com/news/iam-outbound-federation-vpc-endpoints-oidc/</link><guid>https://aiagentaccesscontrol.com/news/iam-outbound-federation-vpc-endpoints-oidc/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AWS announced on 25 September 2026 that IAM outbound identity federation supports interface VPC endpoints for OIDC discovery and JWKS key endpoints through</description></item>
<item><title>New AgentCore Runtime keeps hardware-enforced session isolation</title><link>https://aiagentaccesscontrol.com/news/agentcore-runtime-v2-session-isolation/</link><guid>https://aiagentaccesscontrol.com/news/agentcore-runtime-v2-session-isolation/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>AWS announced on 18 September 2026 the next generation of AgentCore Runtime, the serverless microVM compute in Amazon Bedrock AgentCore. It keeps hardware-</description></item>
<item><title>Policy Troubleshooter can now explain access denials for agents</title><link>https://aiagentaccesscontrol.com/news/policy-troubleshooter-agent-access-denials/</link><guid>https://aiagentaccesscontrol.com/news/policy-troubleshooter-agent-access-denials/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Google Cloud announced on 15 September 2026 that Policy Troubleshooter supports troubleshooting access for agent identities. You can check IAM allow polici</description></item>
<item><title>Google agent platform shell and computer use sandboxes reach GA with VPC-SC</title><link>https://aiagentaccesscontrol.com/news/agent-platform-sandboxes-vpc-sc-psc-ga/</link><guid>https://aiagentaccesscontrol.com/news/agent-platform-sandboxes-vpc-sc-psc-ga/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Google Cloud made Computer Use and Shell sandboxes in Gemini Enterprise Agent Platform generally available on 9 September 2026. The release adds VPC Servic</description></item>
<item><title>Google IAM unified access policies for agents reach general availability</title><link>https://aiagentaccesscontrol.com/news/agent-platform-iam-access-policies-ga/</link><guid>https://aiagentaccesscontrol.com/news/agent-platform-iam-access-policies-ga/</guid><pubDate>Mon, 16 Nov 2026 00:00:00 GMT</pubDate><description>Google Cloud made IAM Unified Access Policies for Gemini Enterprise Agent Platform generally available on 31 August 2026. Agent Gateway uses Identity-Aware</description></item>
</channel></rss>
