Skip to content
Merged
Show file tree
Hide file tree
Changes from 3 commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
20 changes: 6 additions & 14 deletions solutions/observability/observability-ai-assistant.md
Original file line number Diff line number Diff line change
Expand Up @@ -32,26 +32,18 @@ The {{obs-ai-assistant}} helps you:

## Requirements [obs-ai-requirements]

The AI assistant requires the following:
To set up or use AI assistant, you need the following:

- An **Elastic deployment**:
* An appropriate [Elastic license](https://www.elastic.co/subscriptions).

- For **{{observability}}**: {{stack}} version **8.9** or later, or an **{{observability}} serverless project**.
* The `Observability AI Assistant: All` {{kib}} privilege.

- For **Search**: {{stack}} version **8.16.0** or later, or **{{serverless-short}} {{es}} project**.
* An [LLM connector](/solutions/security/ai/set-up-connectors-for-large-language-models-llm.md).

- To run {{obs-ai-assistant}} on a self-hosted Elastic stack, you need an [appropriate license](https://www.elastic.co/subscriptions).

- An account with a third-party generative AI provider that preferably supports function calling. If your AI provider does not support function calling, you can configure [AI Assistant settings](../../solutions/observability/observability-ai-assistant.md#obs-ai-settings) to simulate function calling, but this might affect performance.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I know function calling is something used in Observability that Security doesn't use. We may want to keep a note that recommends a model that supports function calling or to configure simulated function calling.

I would only think this wouldn't be necessary if all of the models we have connectors for support function calling. @viduni94 maybe you would have some insight into this?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Good catch, thanks. Curious to hear further input but sounds like adding a note makes sense.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

reached out to Viduni directly

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

After my conversation with Viduni, I don't think we need to add a note. Recent proprietary models all support function calling, and if the selected model doesn't support function calling, we use simulated function calling so assistant still works.


- The free tier offered by third-party generative AI provider may not be sufficient for the proper functioning of the AI assistant. In most cases, a paid subscription to one of the supported providers is required.

Refer to the [documentation](kibana://reference/connectors-kibana/gen-ai-connectors.md) for your provider to learn about supported and default models.

* The knowledge base requires a 4 GB {{ml}} node.
* To use knowledge base: a 4 GB {{ml}} node.
- In {{ecloud}} or {{ece}}, if you have Machine Learning autoscaling enabled, Machine Learning nodes will be started when using the knowledge base and AI Assistant. Therefore using these features will incur additional costs.

* A self-deployed connector service if you're using [content connectors](elasticsearch://reference/search-connectors/index.md) to populate external data into the knowledge base.
* To use [content connectors](elasticsearch://reference/search-connectors/index.md) to add external data to knowledge base: A self-deployed connector service.

## Manage access to AI Assistant

Expand Down
4 changes: 2 additions & 2 deletions solutions/security/ai/ai-assistant.md
Original file line number Diff line number Diff line change
Expand Up @@ -29,9 +29,9 @@ The Elastic AI Assistant is designed to enhance your analysis with smart dialogu
::::{admonition} Requirements
* {applies_to}`stack: ga` An [Enterprise subscription](https://www.elastic.co/pricing).
* {applies_to}`serverless: ga` An {{sec-serverless}} project with the [EASE or Security Analytics Complete feature tier](/deploy-manage/deploy/elastic-cloud/project-settings.md).
* To use AI Assistant, the **Elastic AI Assistant : All** and **Actions and Connectors : Read** [privileges](/deploy-manage/users-roles/cluster-or-deployment-auth/kibana-privileges.md).
* To use AI Assistant, the **Elastic AI Assistant: All** Security [privilege](/deploy-manage/users-roles/cluster-or-deployment-auth/kibana-privileges.md) and the **Actions and Connectors: Read** management [privilege](/deploy-manage/users-roles/cluster-or-deployment-auth/kibana-privileges.md).
* To set up AI Assistant, the **Actions and Connectors : All** [privilege](/deploy-manage/users-roles/cluster-or-deployment-auth/kibana-privileges.md).
* A [generative AI connector](/solutions/security/ai/set-up-connectors-for-large-language-models-llm.md), which AI Assistant uses to generate responses.
* An [LLM connector](/solutions/security/ai/set-up-connectors-for-large-language-models-llm.md), which AI Assistant uses to generate responses.
* A [machine learning node](/explore-analyze/machine-learning/setting-up-machine-learning.md).
::::

Expand Down
Loading