Documentation Index

Fetch the complete documentation index at: https://docs.lucidity.cloud/llms.txt

Use this file to discover all available pages before exploring further.

AI Analytics in Reports

  • Published on May 5, 2026
  • 3 minute(s) read
Prev Next

AI Analytics in Reports

AI Analytics is a natural language interface inside the Lucidity dashboard that lets you ask questions about your storage data and get instant answers, instead of exporting CSVs or waiting for a custom report.

The feature is currently in beta and rolled out for customers gradually.

Where to find it

Sign in to the Lucidity dashboard, go to the Reports section, and click the Explore with AI banner at the top of the page.

What you can ask

You can ask questions in plain English across the following data domains:

  • Savings. Storage savings totals, savings trends, savings by account or environment, disk pool metrics, and utilization.
  • Account details. Account-level summaries, instance counts, and cloud entity information.
  • Environment mapping. How your instances are tagged across production and non-production environments.
  • Operations. Disk expand and shrink history, onboarding status, agent installation status and failure logs, and file system operation history.

Example questions

  • What were my savings in the last quarter?
  • Show me the savings trend for the last 6 months.
  • How many VMs have we onboarded this month?
  • What are the common causes of agent installation failures?
  • Compare savings between production and non-production for Q1.

What you get back

A typical response includes a natural language summary, a data table where applicable, and follow-up suggestions you can click to explore further. You can also continue the conversation with follow-up questions in the same session.

Coming soon

The following capabilities are on the immediate roadmap:

  • CSV export of any query result
  • Like and dislike feedback on responses to help us improve answer quality
  • Assessment data available alongside reporting data

How your data is used by AI

Lucidity takes data privacy and security seriously. This section explains exactly how your data flows when you use AI Analytics.

Your data is never used to train AI models

When you ask a question, your prompt and the resulting answer are processed by an AI model to generate a SQL-like query against your Lucidity data. None of this content is used to train or improve any AI model, including the underlying foundation models.

Where the AI runs

We use enterprise-grade AI services from the same cloud provider that hosts your Lucidity data:

  • AWS production environments use Amazon Bedrock as the AI layer.
  • Azure (PLS) environments will use Azure OpenAI when that rollout is complete.

Both services are SOC 2 Type 2 compliant. Both contractually guarantee that customer prompts and completions are not used to retrain their foundation models.

Amazon Bedrock guarantees (AWS)

  • No data sharing. Your data is never used to train foundation models, including those from Anthropic, Meta, Mistral, Stability AI, or Amazon Titan.
  • No retention. Bedrock does not store or log your prompts and completions in service logs.
  • Encryption. Your data is encrypted at rest and in transit.
  • Private customization. Any fine-tuning would create a private copy of the base model that is exclusive to Lucidity, ensuring your data remains segregated.

Azure OpenAI guarantees (Azure PLS)

  • No training on customer data. Your inputs (prompts) and outputs (completions) are not used to improve or retrain OpenAI or Microsoft base models.
  • Data isolation. Your data is not available to other customers or to OpenAI.
  • Data residency. Processing can be pinned to specific Azure regions for regional compliance.
  • Compliance coverage. Azure OpenAI is covered by Microsoft's overarching SOC 2 Type 2 attestation, and additionally meets HIPAA, GDPR, ISO 27001, and FedRAMP High.
  • Customer control. You retain ownership of your data and can request deletion at any time.

Tenant isolation inside Lucidity

In addition to the AI provider guarantees above, Lucidity enforces strict tenant isolation in its own application layer:

  • Every query is automatically scoped to the authenticated user's account on the server side. Cross-customer data access is not possible by design.
  • The AI Analytics feature is read-only. Any question that could be interpreted as a write, update, or delete operation is rejected. To make changes to your environment, use the Lucidity dashboard or APIs.
  • Queries are limited to the data you would see in your existing Lucidity dashboard. The AI cannot reach data outside your account boundary.

What is and is not stored

  • Your questions and the answers are shown only in your active session. There is no persistent chat history across sessions today.
  • Lucidity logs aggregate, non-sensitive metadata about feature usage (for example, that a query was submitted, response time, and whether the query succeeded). This is used to monitor performance and reliability, and is consistent with Lucidity's existing product analytics practices.
  • Lucidity does not train any model on your usage, your queries, or your data.

Questions or concerns

If you have additional questions about how your data is handled, please reach out to your Customer Success contact or contact Lucidity Support through the Support Request workflow in the dashboard.

Related articles