About Logiqc AI

Complimentary credits

Logiqc AI runs on a complimentary, plan-based credit allowance included with Pro and Enterprise plans. You'll receive these credits on the 1st of every month, with the amount based on your plan. If you use more than your allowance, you can buy additional credits.

How Logiqc AI works

On the Incident Details form, write a description of what happened, then select the fill-in button to have AI populate the rest of the form:

• Write a useful description. The more detail you give it, the better job it will do.
• Select Fill in this form using AI to have AI suggest information for supported fields.
• Review fields marked with a sparkle icon , and tick the validation checkbox before submitting.

The fill-in button sits near the description field.

AI security

Logiqc currently uses GPT-4o-mini which Microsoft does not make available as local deployment type. This means that the actual location of inference (where the prompt is processed by AI models) is done within Microsoft's global ecosystem. All data at rest is stored strictly in Australia

When using our AI feature, only the data strictly necessary to generate the requested response is transmitted to the Azure OpenAI Service for processing (inference). This includes:

• Form Content & Inputs: Text entered into form fields, selected dropdown values, category selections, and specific configuration settings (such as Severity settings).
• Contextual Data: Any underlying operational or background context required by the prompt to return an accurate result.

Data Handling, Privacy, and Storage Boundaries

Prompts are not de-identified, pseudonymised, masked, or redacted. Additionally, any supporting select fields with dynamic content is also sent as readable text for the purposes of inference. Prompts are discarded after this use and files uploaded or attached to an incident report are not submitted to OpenAI for inference.

• Inference (Transient Processing): Transmitted data is processed in-memory solely to generate the response. It is discarded immediately after the inference request is completed.
• No Model Training: Customer data submitted via Azure OpenAI is not used to train, retrain, or improve Microsoft's foundational models or any third-party models.
• Data Residency at Rest: Generated AI outputs and system records stored at rest remain located strictly within Microsoft Azure data centres in Australia.
• Abuse Monitoring: Transferred data is processed in accordance with Azure OpenAI enterprise security standards. Access to any temporary monitoring logs is strictly controlled by Microsoft security operations under regulated compliance frameworks, ensuring customer data remains isolated and protected.

Audit History & Traceability

Logiqc keeps a record of who used the AI, how many tokens and credits were used, when, and on what type of activity in the platform. We do not keep 'A/B' records showing what suggestions were made by AI and how the user edited those records.

We do record which Incidents have been submitted with generative AI.

Azure Abuse Monitoring & Review

Azure abuse monitoring is designed to detect and mitigate harmful or non-compliant use of the service. It uses content classification, pattern detection, and review processes to identify potential violations of the Code of Conduct.

By default, prompts and completions flagged as potentially harmful are reviewed using automated AI-based analysis. Human review is not automatically enabled unless automated checks fail to meet confidence thresholds or are unavailable. In such cases, authorized Microsoft employees may review flagged content under strict security controls.

Microsoft only allows disabling of Azure abuse monitoring under special consideration. Logiqc has not sought such permission.

Administrator controls

A quick reference for the Logiqc AI actions available to administrators.

Getting help

Most users complete AI fill-in without any trouble. The most common issues, and how to resolve them, are below.