# AI safety does not stop at the model

DevFeed: [AI safety does not stop at the model](<https://devfeed.tech/articles/ai-safety-does-not-stop-at-the-model-35705.md>)

Original publisher: [Read original article](<https://temporal.io/blog/ai-safety-does-not-stop-at-the-model>)

Author: Samar Abbas

Published: 2026-09-15T00:00:00Z

Content type: opinion

Language: en

Sources: [Temporal Blog](<https://devfeed.tech/sources/temporal-blog.md>)

Topics: [ai safety](<https://devfeed.tech/topics/ai-safety.md>), [Agent Harness](<https://devfeed.tech/topics/agent-harness.md>), [reliability](<https://devfeed.tech/topics/reliability.md>), [retry](<https://devfeed.tech/topics/retry.md>), [Security](<https://devfeed.tech/topics/security.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agent-harness](<https://devfeed.tech/tags/agent-harness.md>), [ai](<https://devfeed.tech/tags/ai.md>), [reliability](<https://devfeed.tech/tags/reliability.md>), [retry](<https://devfeed.tech/tags/retry.md>), [safety](<https://devfeed.tech/tags/safety.md>), [security](<https://devfeed.tech/tags/security.md>), [temporal-voices](<https://devfeed.tech/tags/temporal-voices.md>)

## AI overview

The article argues that AI safety extends beyond model behavior to the application layer, where companies must control agent authority and enforce approvals, policies, and credential limits. It emphasizes that these controls must remain reliable through crashes, timeouts, and retries, and presents Temporal Agent Harness as an execution-layer control point around an agent SDK.

## Source excerpt

Auditing what an agent did is only half the job. Companies also have to control what agents may do, and make those limits hold when systems fail.