At Imagicle, we believe Artificial Intelligence (AI) can be leveraged to power better, more accessible communication insights for all. We also recognize that by applying this technology, we have a responsibility to mitigate potential harm. That is why Imagicle develops its AI-powered features in accordance with its responsible AI principles [the “Framework”], which are based on the principles of Transparency, Fairness, Accountability, Privacy, Security, and Reliability. Imagicle translates these principles into product development requirements that form part of the product development lifecycle, alongside our Security by Design and Privacy by Design processes.
Accordingly, Imagicle Advanced AI Analytics features that leverage AI are built with transparency, fairness, accountability, privacy, security, and reliability at their core. Each feature powered by AI undergoes an internal AI impact review — an assessment of how the technical underpinnings of the functionality measure against these principles.
Imagicle Advanced AI Analytics AI features are built with these principles at the center of how Imagicle delivers the AI-powered technology. This Technical Note describes more information about Imagicle Advanced AI Analytics AI features, and how Imagicle responsibly leverages AI to deliver the functionality.
Feature Overview
CDR Collection and Ingestion
The CDR Collection and Ingestion feature automatically collects Call Detail Records (CDRs) from the customer’s calling platform through an Imagicle UCX Suite Cloud instance connected to Advanced AI Analytics, then normalizes and ingests them into the Advanced AI Analytics data store. CDRs contain call metadata — such as calling and called numbers, date and time, duration, direction, and routing information — and do not include the audio content of conversations. The ingested records form the basis for all downstream dashboards, reports, and cost analyses.
Natural-Language Prompt Interface
The Natural-Language Prompt Interface allows end users to describe, in plain language, the dashboard or report they want to obtain. An AI large language model interprets the request and translates it into the corresponding queries, filters, visualizations, and layout, so users can build analytics content without technical knowledge of the underlying data model. Users can iteratively refine the result through follow-up prompts and review the output before relying on it.
AI-Powered Dashboard Creation
The AI-Powered Dashboard Creation feature builds interactive dashboards over ingested CDR data based on the user’s natural-language description. The feature selects the relevant metrics, filters, and chart types — such as call volumes, traffic trends, answered and missed calls, or busiest hours — and assembles them into a dashboard that users can save, share, and further customize.
AI-Powered Report Generation
The AI-Powered Report Generation feature produces structured reports on calling activity from a natural-language description provided by the user. Reports can aggregate CDR data by user, department, site, or time period, and can be exported and scheduled for recurring delivery, reducing the manual effort required to produce periodic telephony and traffic reporting.
Outbound Call Cost Assignment
The Outbound Call Cost Assignment feature assigns a cost to outbound calls and attributes it to the relevant user, department, or organization, based on customer-configured tariff plans and organizational structure. This enables cost accounting, budget monitoring, and charge-back and bill-back scenarios across the organization. Cost assignment is performed by deterministic rules and does not rely on AI models.
Model Overview
Introduction
Imagicle Advanced AI Analytics AI features are powered by large language model (LLM) technologies applied to Call Detail Record (CDR) data. The AI features — including prompt interpretation and dashboard and report generation — are accessed through OpenRouter, an AI routing service that proxies requests to a range of underlying third-party models and providers. These services are operated within a cloud environment configured by Imagicle. For more information about the underlying third-party services, see the OpenRouter transparency documentation referenced below.
Model Architecture
Imagicle Advanced AI Analytics leverages large-language-model technologies to interpret natural-language user prompts and translate them into structured queries, visualizations, and report layouts applied to ingested CDR data. The large-language-model features are accessed through OpenRouter, which acts as an LLM gateway and routes requests to a broad range of underlying third-party models and providers. The specific model used is selected by Imagicle, the current configuration uses the OpenAI GPT model family, and Imagicle may change the model over time as needs evolve. Deterministic, non-AI components perform CDR collection, ingestion, and outbound call cost assignment. Imagicle does not develop the underlying foundation models; it configures and orchestrates them to deliver the Advanced AI Analytics functionality.
Imagicle Advanced AI Analytics is a cloud-native application hosted on Amazon Web Services (AWS).
Usage Guidelines
Imagicle Advanced AI Analytics AI features operate on Call Detail Records collected from the customer’s calling platform and are available as a cloud-native service. The features are intended for use by authorized administrators, managers, and analysts for purposes such as traffic analysis, capacity planning, and cost accounting, and are not intended to make automated decisions about individuals without human review.
Model Inputs and Outputs
CDR Collection and Ingestion
A typical input is the stream of Call Detail Records produced by the customer’s calling platform and forwarded to Advanced AI Analytics by the connected Imagicle UCX Suite Cloud instance. A typical output is a normalized set of call records — including calling and called party, date and time, duration, direction, and routing information — stored and made available for dashboards, reports, and cost analysis.
Natural-Language Prompt Interface
A typical input is a natural-language description of the dashboard or report the user wants to obtain, together with metadata describing the available CDR fields and metrics. A typical output is a structured definition of the requested dashboard or report — the queries, filters, and visualizations to be applied — which the platform then renders over the ingested data.
AI-Powered Dashboard Creation
A typical input is the structured definition derived from the user’s prompt, applied to the ingested CDR data. A typical output is an interactive dashboard combining the relevant metrics, filters, and chart types, which users can review, save, share, and refine through further prompts.
AI-Powered Report Generation
A typical input is the structured definition derived from the user’s prompt, applied to the ingested CDR data. A typical output is a structured report aggregating calling activity — for example by user, department, site, or time period — which can be exported or scheduled for recurring delivery. Users are encouraged to review generated reports for accuracy before relying on them.
Outbound Call Cost Assignment
A typical input is an outbound call record together with the customer-configured tariff plans and organizational structure. A typical output is a cost assigned to the call and attributed to the relevant user, department, or organization, presented in dashboards and cost reports. This feature is deterministic and does not rely on AI models.
Data Sources for Training and Evaluation
Imagicle does not use customer content to fine-tune or train the foundation models used to deliver Advanced AI Analytics AI features. Customer CDR data and user prompts are processed solely to generate the requested output (dashboards, reports, and cost analyses) for the customer.
Where Advanced AI Analytics relies on a third-party AI service, requests are routed through OpenRouter to the underlying model providers. These providers represent that they do not use customer data submitted through the service to train or improve their foundation models, and do not retain customer data passed to the service beyond what is necessary to provide the functionality. For more information about OpenRouter’s data, privacy, and security practices, see the OpenRouter transparency documentation referenced below.
The underlying foundation models are trained by their respective providers on large datasets compiled from a wide range of publicly available written and spoken material. This training data encompasses a diverse array of topics, languages, and styles to help the models develop a broad understanding of language and its contexts. For further information, see the provider documentation referenced below.
Model Evaluation and Performance
Imagicle periodically assesses the AI features to maintain and improve both the performance and accuracy of the generated output. Humans participate in the review, testing, and quality assurance associated with the deployment of the underlying models. Dashboards and reports generated from natural-language prompts are probabilistic and may contain errors or misinterpretations of the user’s request; accordingly, the features are designed to support — not replace — human review, and users can inspect and refine the generated output before relying on it.
Safety and Ethical Considerations
AI models may produce inaccurate, incomplete, or unintended output if such patterns exist in the input or the model. To mitigate this, the third-party AI services used by Advanced AI Analytics include content-filtering and safety controls, and Imagicle constrains how the models are used to the specific, narrow tasks of interpreting user prompts and generating dashboard and report definitions over CDR data.
In addition to these technical mitigations, Imagicle has taken further steps to support safe and ethical use. Imagicle’s third-party AI vendors undergo vendor reviews that include security and safety assessments. Output generated by Advanced AI Analytics features is visible only to authorized users within the customer’s organization, unless shared by those users voluntarily. Please contact Imagicle if you have concerns or feedback about your experience with Advanced AI Analytics.
Fairness
To understand how the foundation models underlying Advanced AI Analytics are trained and evaluated for fairness, please reference the transparency documentation of the relevant AI service provider (OpenRouter), referenced below. Imagicle supports natural-language prompts in multiple languages and reviews feature output to help identify and reduce systematic disparities in performance.
Privacy and Security
Imagicle Advanced AI Analytics processes Call Detail Records, which contain personal data such as telephone numbers and, where configured by the customer, user, department, and organizational identities. Customers remain the controllers of the CDR data they choose to collect and analyze and are responsible for informing data subjects and for configuring collection and retention in line with applicable law and regulation.
CDR data is encrypted at rest, and access to dashboards, reports, and cost information is restricted to authorized users. Imagicle addresses the security of customer data in its security and data-processing documentation, available on request. Where third-party AI services are used, the provider (OpenRouter) represents that it does not access, monitor, or store customer data beyond what is necessary to provide the service.
Advanced AI Analytics is built on cloud services and third-party open-source and commercial components, which are managed under Imagicle’s secure development lifecycle, including dependency tracking, patching, and vulnerability management. The component inventory changes as the product evolves; a current list can be provided on request.
Data Location
Imagicle Advanced AI Analytics is delivered from regional data centers, and the location in which customer data is stored is determined by the region selected when the customer is onboarded. Customers onboarded in the European region — including customers located in the European Union and in the Middle East — are hosted in a data center located in the European Union. Customers onboarded in the United States are hosted in a data center located in the United States.
Call Detail Records, dashboards, reports, and cost data are stored in the data center of the customer's assigned region. Where Advanced AI Analytics relies on third-party AI services, as described in the Model Overview section, processing by those services is governed by the providers' data-processing terms. For more information about data locations and sub-processors, see Imagicle's security and data-processing documentation, available on request.
Updates and Maintenance
Ongoing Imagicle product updates and release notes describe product changes and improvements to the user experience. Imagicle may also use customer email notifications and its support and community resources to inform customers of changes to Advanced AI Analytics AI features.