AI Studio is available on Starter, Advanced, Enterprise, and Enterprise+ plans. Visit asana.com/ai to learn more about the launch of Asana AI Studio.
This article answers frequently asked questions about AI Studio - from how to get started to how permissions, data access, and model selection work. 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.
To access Asana AI Studio, follow the steps detailed in this article.
Yes. You can take the AI Studio Foundations skill badge in the Asana Academy to learn how to build smart workflows and write effective instructions.
We have a dedicated AI Studio Forum section. This is a space designed for discussing AI Studio, sharing tips and use cases, and asking questions.
You can @mention a team in Asana in the same way as you would mention a task or project. This allows the LLM to see the names of up to 200 members of the team and to see who the team admins are. This means it's possible to use AI to add members of a team as collaborators on a task, or assign a task to a team admin.
Instructions and conditions serve different purposes in a smart workflow:
Instructions are essential—they tell Asana AI how to perform a task by providing context, goals, and guidelines.
Conditions are optional—they determine whether an action should happen based on certain criteria.
Instructions guide Asana AI’s behavior in a smart workflow. They provide necessary context, goals, and details on how Asana AI should complete a task. Every AI-powered workflow must have instructions.
Conditions work like "if-then" statements. After a trigger event, the rule checks if a condition is met before proceeding with an action. If the condition isn’t met, the action won’t occur. However, not all workflows require conditions.
A conditional check by AI is similar to a regular condition, but instead of using fixed criteria, Asana AI evaluates the condition itself. This allows for greater flexibility, as Asana AI can assess more complex or nuanced situations.
AI Studio currently offers a choice of six large language models for use in your smart workflow. The right model to use depends on the type of work to be done and the potential cost.
High cost. Great at writing documents and most tasks.
Very high cost. Best at writing documents and following complex guidance.
High cost. Good at writing documents and most tasks.
Medium cost. Best for complex tasks and following detailed guidance.
Very low cost. Best for most tasks.
Medium cost. Good at summarizing and classifying information.
We’ve recently made updates to the models available in AI Studio. To ensure better performance and a consistent experience, we have upgraded all rules using older models. The following models are no longer available for rules: Claude Sonnet 3.5, Claude Opus 4, o1, o3-mini, GPT-4o, GPT-4o mini. On October 20, 2025, rules using legacy models were automatically migrated as described below:
Improved performance, same cost: The upgraded models are more intelligent and also consume the same or fewer credits, so your overall credit usage will not increase.
Automatic transition: You do not need to do anything to enable this transition. On October 20, 2025, rules using legacy models were automatically migrated for you as described above.
No, there isn’t a limit.
The smart workflow can only access projects and tasks that the builder has permission to see. If the builder doesn’t have access to a project, they can’t set up Asana AI to interact with it. Likewise, if the builder can't view certain tasks, Asana AI won’t be able to see or analyze them when the smart workflow runs.
No, guests can’t be builders.
Yes, customers on Asana’s HIPAA offering can enable AI Studio. For more information on Asana AI and its data protections, please see Asana AI FAQ.
When building a smart workflow in AI Studio, you can dictate the data that Asana AI has access to. This gives you control over exactly what data is handled by Asana AI. By default, an AI-powered rule can access two things:
If a task is multi-homed in several projects, the rule cannot identify the other projects unless they are explicitly mentioned in the instructions.
You can give Asana AI access to additional data in AI Studio by configuring instructions in the following ways:
For more information, visit the article: Create smart workflows with AI Studio.
An input task refers to the task that directly triggers a rule. You have two options under the Data to use from tasks section of the AI instructions:
Task data is the default, non-configurable set of data that may be referenced along with any task. InThis data includes:
*For Asana AI to see attachments, you must check the Include attachments added to tasks checkbox.
If you choose to include activity and linked tasks in addition to the task data and comments, the following are sent to Asana AI:
Web access is currently in early access and only available for Advanced (annual), Enterprise, and Enterprise+ customers.
Web access is a feature that allows Asana AI to gather real-time information from public websites. This includes both conducting web searches and reading content from specific user-provided links.
Asana AI will read:
Admins can enable web access from the Asana AI section in the admin console. Once enabled, users can customize access per workflow in AI Studio.
Yes, both Anthropic and OpenAI models support web access.
No. This feature is not available in HIPAA-compliant workspaces due to limitations set by Asana's AI providers.