Davis CoPilot Frequently Asked Questions
If you have any questions related to Davis CoPilot, please start by seeing if they have already been answered in the FAQ.
Does my account need to be on DPS to use Davis CoPilot?
No. Davis CoPilot is available on any SaaS environment running the latest Dynatrace, irrespective of your licensing model.
Will Davis CoPilot be licensed?
No. There is no licensing associated with our current generative AI functionality. However, even though Davis CoPilot is not charged for, all queries that are executed by Davis CoPilot are subject to licensing consumption according to your existing licensing agreement. If you are concerned about the cost of auto-executing generated queries, you can choose to generate DQL only, without executing it. For more information, see Quick analysis with Davis CoPilot.
How will Davis CoPilot impact my license consumption?
Davis CoPilot itself has no impact on your license consumption. However, all queries that are executed by Davis CoPilot are subject to license consumption according to your existing license agreement. For more information, see Will Davis CoPilot be licensed?.
Is my data used to train Davis CoPilot?
No. Customer data and customer prompts are not used for training. Davis CoPilot is based on a retrieval augmented generation (RAG) approach, which means that data and additional context is used only to enrich prompts. The model does not learn from this. Customer data isn't used to automatically fine-tune, train, or improve any models or services, either by Dynatrace or by Microsoft.
Is my data used to train the Davis CoPilot model for other customers?
No, customer data is not used to train Davis CoPilot at all. There is no risk of customer data being shared between environments.
Does Davis CoPilot collect my data?
For the basic Davis CoPilot functionality, there is no access to customer data. The generic query generation relies only on public domain knowledge (semantic dictionary, DQL documentation, etc.) and an extensive, curated dataset that is managed by Dynatrace. The only customer context that Davis CoPilot has access to is the user's prompt.
However, if you enable environment-aware queries, Davis CoPilot will create a semantic index that is stored in the Dynatrace platform. Creating a semantic index allows Davis CoPilot to generate more accurate queries that identify and reference relevant entities, events, and metrics. This lets you run deeper and more complex data analysis by asking about the specifics of the data in your environment. Keep in mind that:
- You have full control over which data in Grail is semantically indexed by Davis CoPilot on the level of Grail data objects and buckets.
- Even when environment-aware queries are enabled, your data is still stored only within your environment, and the data fragments that are used to enrich your prompts are not used to train or fine-tune the model.
What are the prerequisites for using Davis CoPilot?
- Your environment must be SaaS and running the latest Dynatrace.
- You must enable Davis CoPilot on your environment.
- You must assign permission to the relevant users or user groups.
For details, see Getting started with Davis CoPilot.
Is Davis CoPilot available on SaaS and Managed?
Davis CoPilot will be available for all Dynatrace SaaS customers using the latest Dynatrace. Both AWS and Azure accounts are supported.
Davis CoPilot is not available for Dynatrace Managed customers.
Where can I use Davis CoPilot?
The Davis CoPilot functionality currently allows you to conduct quick data analysis in Notebooks and Dashboards by generating DQL based on your natural language prompts.
How can I provide feedback?
You can provide feedback directly in the Notebooks or Dashboards apps, or join our dedicated Community user groupļ»æ. To learn how to give feedback through Dashboards and Notebooks, see Quick analysis with Davis CoPilot.
Can Davis CoPilot handle questions in foreign languages?
Yes. Davis CoPilot does respond to commands and questions in languages other than English.
What is the difference between Davis and Davis CoPilot?
The Davis Hypermodal AI refers to Dynatrace's composite AI. This means that our AI offering consists of different types of AI:
- Causal AI
- Predictive AI
- Generative AI (Davis CoPilot)
Each of these types of AI is built on a different AI model and addresses different use cases:
- Causal AI is dedicated to finding root-cause and impact by using Smartscape topology and environment-specific observability data, such as metrics, traces, logs, events, and user sessions.
- Predictive AI uses timeseries data to train models to detect and alert on anomalies and to predict future values and trends.
- Generative AI generates content (for example, answers, queries, documents, and suggestions) based the user input (prompt) combined with environment data and configuration.