# Why AI Agents Need Dedicated Identity and Access Controls

DevFeed: [Why AI Agents Need Dedicated Identity and Access Controls](<https://devfeed.tech/articles/why-your-ai-agent-has-an-identity-problem-23743.md>)

Original publisher: [Read original article](<https://cockroachlabs.com/blog/ai-agent-identity-security>)

Author: Quentin Packard

Published: 2026-07-17T00:00:00Z

Content type: opinion

Language: en

Sources: [Cockroach Labs](<https://devfeed.tech/sources/cockroach-labs.md>)

Topics: [AI Agent](<https://devfeed.tech/topics/ai-agent.md>), [identity and access management](<https://devfeed.tech/topics/identity-and-access-management.md>), [Security](<https://devfeed.tech/topics/security.md>), [Authorization](<https://devfeed.tech/topics/authorization.md>), [audit trail](<https://devfeed.tech/topics/audit-trail.md>)

Tags: [ai-agent](<https://devfeed.tech/tags/ai-agent.md>), [apis](<https://devfeed.tech/tags/apis.md>), [audit-trail](<https://devfeed.tech/tags/audit-trail.md>), [identity-and-access-management](<https://devfeed.tech/tags/identity-and-access-management.md>), [permission](<https://devfeed.tech/tags/permission.md>), [security](<https://devfeed.tech/tags/security.md>)

## AI overview

The article argues that production AI agents need dedicated identities and access controls because shared service accounts, broad permissions, and incomplete audit trails make it difficult to determine what agents accessed and why. Traditional IAM was designed around human actors and does not fully address autonomous software processes that act rapidly, access many records, call external APIs, or delegate work to other agents.

## Source excerpt

The agent your team just shipped to external users has the ability to read customer records, execute transactions, and call external APIs.