AWS

Amazon Web Services (AWS) is Amazon's cloud computing platform, offering businesses rented access to servers, data storage, and a large catalog of AI services such as Amazon Bedrock, which gives companies access to foundation models from several providers. For enterprise AI governance, AWS matters because many organizations run their models, training data, and customer information on its infrastructure, so its contract terms, data residency options, and shared-responsibility security model directly shape a company's own compliance obligations. Regulators and auditors will often look at how a cloud provider's controls fit into the customer's overall risk program.

1 item

ResearchGlobal2026-10-08

AWS AgentCore's Recurring Credential Flaws Expose a Structural Agentic Risk

Security researchers at Palo Alto Networks Unit 42 and Zenity Labs documented multiple credential theft vulnerabilities in AWS AgentCore throughout 2025 and 2026. Successive patches closed individual attack paths without resolving the underlying problem: broad access rights that enable agents to function also expose live credentials and internal systems. The findings carry direct implications for how enterprises govern agentic AI access, vendor shared-responsibility assumptions, and least-privilege controls.

AWS agentic AI credential security least privilege Palo Alto Networks enterprise risk cloud security