SW1 for SolarWinds Observability SaaS - FAQ

This page may be updated from time to time, so check back regularly for the latest version of these FAQs.

What is SW1?

SW1 is the SolarWinds AI agent built into multiple SolarWinds products, including our SolarWinds Observability SaaS. It lets users ask natural-language questions inside the product and receive brief explanations, insights, and recommended next steps based on the data they already have access to.

How do you set it up?

For SaaS deployments, SW1 is available within your SolarWinds Observability SaaS tenant. It is enabled at the tenant level by administrators. No additional infrastructure or configuration is required. Admin users can enable or disable SW1 from the AI Settings within the product. Once enabled, the SW1 icon appears at the top right corner of the application.

What data is being sent?

When a user submits a request, a subset of relevant observability data - such as metrics, alerts, logs, or information relevant to the context - may be sent to an external large language model (LLM) to generate a response.

Only data the user already has access to within the product is included. Sensitive data may be included in limited amounts where necessary for SW1 to provide an accurate response. We do have PII masking in place, which looks for standard patterns to identify PII. Masking takes place prior to any data being sent to an LLM.

How is that data being sent?

The relevant data is encrypted in transit. For SaaS deployments, the request is processed within the SolarWinds platform through the agentic and LLM gateway layers. The LLM gateway routes the request to AWS Bedrock, where it is processed, and a response is returned to the product UI.

For SaaS deployments, the high-level request path is shown in the figure below.

How long is my data stored?

SW1 prompts, outputs, and related logs are retained according to SolarWinds privacy and retention policies, detailed at solarwinds.com/legal/privacy.

What controls are in place to ensure my data privacy?

Three key controls apply: existing role-based access controls govern all in-product activities; data exchanged with external model providers is encrypted in transit; and administrators can disable SW1 at the tenant level.

Is my data used to train AI models?

No. Customer data is not used to train underlying AI models for language model providers. Interactions with model providers are designed to be stateless at the LLM layer.

Does SW1 change anything in my environment without my consent?

No. SolarWinds follows a human-in-the-loop approach requiring user consent before environment changes. SW1 operates only within existing user permissions.

Where is SW1-related data stored (data residency)?

AI-related data that SolarWinds retains (prompts, context, outputs, logs) is stored in the same region(s) as the customer's SaaS instance.

Is SW1 covered by SOC 2 / security certifications?

Yes.