Agent One brings together AI-powered experts—referred to as agents—designed to support marketers and customer engagement teams with scalable, personalized interactions. These agents operate through emotionally intelligent conversations and autonomous decision-making, helping users solve complex engagement challenges efficiently.
Each agent is trained using deep industry knowledge and built to address specific use cases across the customer journey. With a focus on precision and speed, Agent One enables teams to drive impactful outcomes and deliver more relevant, timely, and adaptive experiences.
Agent One reflects the Insider’s continued evolution in incorporating both Generative AI and Agentic AI. The solution is intentionally designed to eliminate the need for heavy customization, allowing teams to avoid costly and time-consuming development cycles.
Unlike generic AI solutions, Agent One leverages Insider’s integrated CDP and metadata infrastructure to enrich customer interactions with continuous intelligence. Every conversation contributes to a deeper understanding of your audience, enabling hyper-personalized engagement that evolves in real time and strengthens long-term loyalty.
Agent One currently includes two specialized AI agents, each focusing on distinct interaction goals from service automation to shopping assistance:
The Support Agent automates customer service conversations with contextual, brand-aligned responses. It resolves FAQs, guides users through troubleshooting, and escalates complex cases to live support, improving efficiency and customer satisfaction. Get started with the Support Agent.
The Shopping Agent enhances digital shopping experiences with real-time, personalized product guidance.
It connects to your catalog, recommendation, and search systems to help users discover, compare, and purchase products seamlessly. Get started with the Shopping Agent.
Agent One Capacity
The table below outlines the current capabilities of the Agents and the communication channels through which they can operate. These details will help you understand the system's scalability and how it can be integrated into your existing support workflows.
Capability | Description | Value |
|---|---|---|
Maximum number of Agents | The total number of Agents that can be created per InOne panel. | 10 |
Supported message input types | The types of messages that Agents can accept. | Text messages only |
Maximum timeout duration for Agents | The longest time an agent can wait for a user’s input before responding. | 90 minutes |
Maximum number of Actions | The highest number of unique Actions (such as API calls and data-collection steps) that can be configured per agent. | 25 (Exit & API actions combined) |
Maximum number of Knowledge bases | The total number of separate knowledge base sets that can be created within a workspace or environment. | 4 |
Maximum number of Knowledge base sources | The maximum number of unique source URLs that can be crawled and connected to a single agent. | 5 |
LLM Sources & GDPR Compliance
The agents utilize a large language model (LLM) provided by OpenAI GPT-4.1, which is accessed securely through Azure OpenAI Service, Microsoft’s fully managed and enterprise-grade deployment platform.
Language Model Source
The underlying AI model powering the Agent’s natural language understanding and generation is GPT-4.1, developed by OpenAI. This model is widely recognized for its strong reasoning capabilities, multilingual fluency, and safe response alignment.
GPT-4.1 is accessed via the Azure OpenAI Service, ensuring:
No model training or fine-tuning occurs on customer data.
No data is retained by Microsoft or OpenAI after processing each request.
All model calls are handled within Azure’s secure infrastructure, enabling region-bound and compliant computation aligned with enterprise and regulatory requirements.
GDPR Compliance
The Agent platform is developed in alignment with key GDPR principles, including data minimization and purpose limitation. As a data processor, Insider implements appropriate safeguards to support GDPR compliance for its enterprise customers. Partners remain responsible for obtaining a valid legal basis, including consent where necessary, from end-users.
Key compliance measures include:
Data minimization: Only information necessary for processing is transmitted to the underlying LLM.
No data retention: All interactions are processed in a stateless manner. Based on available provider documentation, neither Microsoft Azure nor OpenAI retains any data exchanged during the model interaction.
These measures help ensure that Agent users can benefit from AI capabilities in a privacy-conscious and compliant manner, in accordance with relevant regulations and guidelines.