Loading
PLATFORM
CAPABILITIES
ASANA AI
PLANS
Company Type
Teams
Industries
Use Cases
LEARN
SUPPORT
SERVICES
Help Center
search-icon

Topics

AI Studio - Human in the loop

In This Article

AI Studio is available for Starter, Advanced, Enterprise, and Enterprise+ customers. Visit asana.com/ai to learn more about the launch of Asana AI Studio.

When building Smart workflows via AI Studio, it is best practice to keep a human in the loop. This ensures that critical tasks—such as reviewing drafts and making strategic decisions—are handled with appropriate care. This article will show you how to add human-in-the-loop checkpoints to your Smart workflows. If you’d like to go further, you can also earn the AI Studio Foundations skill badge in the Asana Academy. You’ll learn how to write effective prompts, plan Smart workflows, and troubleshoot common issues.

Related articles:

Why human oversight matters

Maintaining human oversight at critical checkpoints ensures:

  • Quality control: Humans can verify that AI decisions and outputs align with relevant standards
  • Contextual understanding: Expert review adds nuance that AI might miss
  • Risk management: Critical decisions benefit from human judgment

To incorporate human oversight, you can set up human in the loop (HITL) checkpoints by adding approval tasks or similar mechanisms to your smart workflow. At this stage, the task can be approved to proceed, rejected to stop the process, or edited and redirected as needed.

Example workflow with a human in the loop

To see this concept in action, here’s an example of how a marketing team integrates a HITL checkpoint into their content creation workflow. This workflow is designed to create and draft social media posts with built-in human oversight to maintain quality and accuracy throughout the process.

The workflow begins with AI handling the initial steps. When a task is added via a form, the AI reviews it and drafts the social media post. Once the draft is complete, the task automatically moves to the "HITL Draft Review" section, where a human reviewer is assigned to ensure the content meets standards and is ready to move onto the next stage.

At this point, the workflow pauses when the approval task is created for Jamie. Nothing progresses until they review and approve the draft, ensuring that only finalized content continues through the process. This checkpoint prevents the team from translating or publishing incomplete drafts, saving time and resources.

Once Jamie has reviewed and approved the draft, the smart workflow seamlessly resumes, moving the task to the "Translation" section where the content is prepared for publication. The HITL checkpoint ensures quality while maintaining the efficiency of the automated flow, providing a brief but essential pause for human review.

Common use cases for HITL checkpoints

HITL checkpoints are versatile and can be applied to a wide range of workflows across industries. 

In marketing, they ensure that content, such as social media posts, ad copy, or campaign drafts, aligns with brand guidelines and messaging standards before publication. 

In customer support, they provide oversight for AI-generated responses to ensure accuracy and appropriateness before reaching customers. 

By embedding these checkpoints, teams can trust that AI is working effectively while maintaining control over key decisions. This approach not only safeguards quality but also helps teams iterate and improve their workflows over time, as feedback from reviewers informs future refinements. HITL checkpoints create a safety net, enabling teams to confidently adopt AI-driven processes without compromising on accuracy or oversight. Click here if you’re interested in reading about our AI principles.

Related articles

Loading
How to build human-in-loop AI | Asana Help Center