Curity looks to reinvent IAM with runtime authorization for AI agents
Briefly

Curity looks to reinvent IAM with runtime authorization for AI agents
"Traditional IAM tools assume applications are accessed by human users or machine identities, governed by a one-time authentication process. Agents, however, operate with ephemeral and complex access patterns that are unpredictable."
"Curity's approach treats agents as a special type of application, using OAuth tokens to not only permit access but also to convey the agent's purpose and intent."
"Access for agents is granted at runtime, with each requested action generating a separate token that specifies the permissions needed for that action."
"In high-risk scenarios, such as transferring funds, human authorization may be required, balancing the need for security with the operational flexibility of agents."
Traditional Identity and Access Management (IAM) tools are designed for human users and machine identities, relying on one-time authentication. However, agents operate differently, with access that is ephemeral and unpredictable. Curity addresses this by treating agents as unique applications, utilizing OAuth tokens not just for access but also to convey the agent's purpose. Access is granted dynamically at runtime, with each action generating a specific token. High-risk actions may require human authorization, ensuring security while allowing agents to function effectively.
Read at InfoWorld
Unable to calculate read time
[
|
]