AI features FAQs

Totara's AI capabilities are designed to enhance content creation, streamline learning experiences, and support goal setting across the platform. As these features continue to evolve, this article answers frequently asked questions about what’s available now, how AI is used in Totara, and what to consider when enabling it on your site.

What AI features are available in Totara?

Totara offers a growing suite of AI features that span writing assistance, image generation, course tagging, informal quiz generation, SMART goal setting, and personalised content recommendations.

Are AI features available to all customers by default?

Yes, however these features depend on having an integration with a large language model (LLM), like OpenAI. Customers can use Totara’s existing OpenAI integration for ease of setup or configure their own integration to a different LLM. Partners can also manage the integration process on behalf of customers.

How do I get access to the Totara AI features?

These features are available as an add-on to your Totara site. To get access to Totara AI features talk to your Totara Partner for more information and to get access.

Is there a cost associated with using AI features?

Yes. AI features come with usage-based costs determined by the external LLM provider. For example, OpenAI charges per token or usage volume, and these costs are billed directly between the customer and the LLM vendor. Totara does not add an additional charge. Cost profiling work is underway to provide clearer guidance in future releases.

Do AI features require customer data to be shared externally?

Totara AI features do not share data across different customers or sites. All AI interactions are scoped to the specific site and user. Memory and content are not accessible by other instances or users, and all actions respect Totara’s AI principles for purposeful and compliant use.

When the OpenAI integration powers Totara AI features, this is using OpenAI’s API, where data is processed securely and not used to train OpenAI’s models. Specifically, when using OpenAI’s API services (as opposed to using ChatGPT via their consumer-facing platform), inputs and outputs are not stored or used for model training. This ensures your data remains private and protected.

You can find further technical assurances in OpenAI’s API data usage policy.

Can AI features be used with a customer’s own LLM?

Yes. While Totara offers a built-in integration with OpenAI, with Totara Partner support, customers can also choose to link AI features to their own LLM setup. This supports a more tailored deployment model.

Note: this will take development effort and only a technical partner will be capable of this.

Can the AI features use additional customer-specific context when generating responses?

Not at this stage. The current AI features do not incorporate broader site context or additional user-specific data beyond the immediate input provided in the prompt. This helps ensure data isolation and simplifies compliance. However, we are actively exploring ways to safely introduce contextual prompts in future releases to enhance output relevance and value.

Are AI features restricted to certain roles?

Yes. All AI features are gated behind defined capabilities, which allows site administrators to control access based on role. This ensures that customers can safely experiment with AI usage, limit access to trained users, and manage usage-based costs effectively. For example, you might enable AI tools only for course creators during an initial rollout phase.

What version of Totara are the AI features available in?

The initial suite of AI features is available from Totara version 19.1 onwards. Customers using earlier versions will need to upgrade to access these capabilities. Future releases will continue to expand the feature set and integration options.

Will Totara support other LLM providers in the future?

We are interested in exploring support for additional LLM providers beyond OpenAI in future releases. While our current implementation focuses on OpenAI due to its ease of integration and broad capabilities, we recognise the importance of flexibility and vendor choice. If a lack of support for another LLM is blocking adoption for you, please share this feedback with your Totara Partner or the Totara team directly.

How does image generation work in Totara?

Totara connects securely to OpenAI’s image generation API to produce images from text-based prompts. The feature sends the prompt to OpenAI, which uses its own models to generate the image and return it to Totara. Totara displays the resulting image but does not alter, enhance, or interpret the image beyond what is returned by OpenAI. Image generation is performed entirely through OpenAI’s systems, following their technical and content guidelines.

How is copyright handled for generated images?

Copyright ownership and usage of generated images follow OpenAI’s Terms of Use and intellectual property policies. Totara does not apply additional copyright filtering, validation, or licensing controls beyond those implemented by OpenAI. Users are responsible for reviewing and complying with applicable copyright and licensing requirements when using generated images in Totara.