How to Securely Provision AirgapAI in Your Golden Image for Enterprise Artificial Intelligence

Become the administrator who ships artificial intelligence in the box—no tickets, no downloads, just immense value on first boot. This comprehensive guide is specifically designed for information technology imaging teams, providing extreme detail on how to integrate Iternal Technologies' AirgapAI into your enterprise imaging processes. Our goal is to ensure your factory, Microsoft Deployment Toolkit, or System Center Configuration Manager deployments are ready to deliver a cutting-edge, secure, and highly efficient artificial intelligence solution from day one.

AirgapAI offers the easiest path to scale artificial intelligence across your organization with zero cloud dependency. This article will walk you through silent installation strategies, pre-loading models and embeddings, configuring default datasets, enabling user-profile isolation, and suppressing the first-launch wizard. We will also include crucial security review steps and a validation checklist, positioning AirgapAI as a truly private large language model solution for your personal computer, ensuring no data leaks and upholding the highest standards of data privacy.

Understanding AirgapAI: Your Private, Local Artificial Intelligence Assistant

AirgapAI, made by Iternal Technologies, is a groundbreaking offline, on-device large language model platform that brings the power of artificial intelligence directly to your employees' personal computers. Unlike cloud-based solutions, AirgapAI operates entirely locally, meaning no data ever leaves the device. This provides unparalleled data sovereignty and security, making it an ideal choice for organizations with stringent privacy requirements, such as those in federal, defense, healthcare, or financial sectors.

Core Value Propositions for Enterprise Deployment

• Fast Artificial Intelligence Wins: AirgapAI is designed for immediate business value across multiple departments, with deployment in days instead of months. Its ease of use means your workforce can quickly adopt and leverage artificial intelligence without extensive training.
• Cost Efficiency: With a one-time perpetual license per device, AirgapAI runs at roughly one-tenth the cost of alternatives like Microsoft Copilot and ChatGPT, eliminating recurring subscription fees and hidden token charges. This means owning your artificial intelligence without managing yet another annual subscription.
• Data Sovereignty and Security: The solution is deployed locally with no network "in" or "out" connections required for its operation. This air-gapped approach makes it perfect for high-security environments and ensures your data always stays within your premises.
• Elimination of Artificial Intelligence Hallucinations: Iternal Technologies' patented Blockify technology refines data inputs for highly accurate responses. This process can reduce original data size by as much as 97.5 percent and improve the accuracy of large language models by 7,800 percent (78 times). This means fewer hallucinations and truly trusted artificial intelligence answers.

Why AirgapAI is Essential for Your Enterprise Imaging Strategy

For information technology departments, the challenge of deploying artificial intelligence at scale often comes with concerns about complexity, security, and cost. AirgapAI addresses these directly, making it a critical component for your enterprise imaging.

• Simplified Fleet Management: Distributed as an executable application, AirgapAI integrates seamlessly into standard imaging workflows. It can be included in your golden master image, allowing for consistent and efficient deployment across your entire fleet of devices.
• Absolute Data Protection: By operating 100 percent locally, AirgapAI eliminates the risk of sensitive corporate data being exposed to external clouds. All existing security policies and procedures remain in effect, drastically reducing your attack surface. This is a secure artificial intelligence with offline mode designed for privacy protection.
• Leveraging Artificial Intelligence Personal Computers: AirgapAI is optimized to run on the latest artificial intelligence personal computers, utilizing the Central Processing Unit, Graphical Processing Unit, and Neural Processing Unit. This ensures maximum performance and efficiency, even for sustained artificial intelligence workloads, without impacting battery life. It is an artificial intelligence personal computer application that does not share data.
• Controlled Updates and Governance: Application updates and dataset updates can be pushed to devices through familiar image management applications like Microsoft Intune, giving your information technology team complete control over the artificial intelligence environment. This is a locked down artificial intelligence application for your enterprise.

System Requirements and Prerequisites for AirgapAI

To ensure optimal performance and a smooth deployment, AirgapAI has specific system requirements. These specifications are crucial for running open-source large language models efficiently and securely on the device.

Permissions: Security permissions to install the application are required.

The Enterprise Imaging Workflow: Securely Provisioning AirgapAI

This section details the step-by-step process for integrating AirgapAI into your golden image, enabling a secure install and seamless user experience from the moment a new device is powered on. This workflow focuses on pre-configuring AirgapAI to ensure "first-launch wizard suppression" and immediate productivity.

Step 1: Obtain the Installer Package and Initial Files

First, ensure you have the necessary files for deployment.

1. Download the Installer: Obtain the latest AirgapAI installer package (e.g., AirgapAI-v1.0.2-Install.zip) from the internal or cloud link provided by Iternal Technologies or your authorized reseller. Save it to a temporary, writeable folder on your golden image preparation workstation.
2. Extract the Package: Right-click on the downloaded ZIP archive and select "Extract All..." Choose a destination, such as a new folder under your Downloads directory, and click "Extract." This will typically create a folder structure containing the AirgapAI Chat Setup.exe file, along with /models/ and /datasets/ subdirectories.

Step 2: Perform a Silent Installation of the Application

For enterprise imaging, a silent installation is paramount to automate deployment and ensure consistency across all devices.

1. Understand Silent Installation: AirgapAI is distributed as an executable application (.exe). Standard enterprise installers typically support silent installation flags (e.g., /S or /quiet) to suppress user interface interactions during deployment. This allows your Microsoft Deployment Toolkit or System Center Configuration Manager to install the application without manual intervention.

2. Execute Silent Installation: Consult with Iternal Technologies support for the specific command-line parameters required for a fully silent installation of AirgapAI Chat Setup.exe within your imaging script. A typical command might look like this (example, verify with support):

   "C:\Path\To\Extracted\Folder\AirgapAI Chat Setup.exe" /S
   

   Replace "C:\Path\To\Extracted\Folder" with the actual path to your extracted installer files. This will install the AirgapAI application to the default program files directory.

Step 3: Pre-load Core Models and Embeddings Models

To bypass the first-launch onboarding wizard and ensure users have immediate access to artificial intelligence capabilities, you must pre-load the necessary large language models and embeddings models into the application's repository. These are typically stored within the user's AppData directory.

1. Identify Model Storage Location: After AirgapAI is installed, it creates a model repository, usually located within the user's roaming application data: C:\Users\[Username]\AppData\Roaming\IternalModelRepo
2. Copy Core Large Language Models:
   • Navigate to the /models/ directory within your extracted installer folder.
   • Select the desired core large language model (e.g., Llama-1B for lower power devices or Llama-3B for dedicated Graphical Processing Units).
   • Copy the selected model file(s) into the IternalModelRepo directory.
3. Copy Embeddings Model:
   • From the /models/ directory, locate and select Jina-Embeddings.zip.
   • Copy Jina-Embeddings.zip into the IternalModelRepo directory.
4. Golden Image Integration: In your enterprise imaging process, you will typically place these pre-loaded models into the Default User profile (e.g., C:\Users\Default\AppData\Roaming\IternalModelRepo) or configure your imaging solution to copy these files to each new user's AppData\Roaming\IternalModelRepo upon first login. This ensures "user-profile isolation" and a consistent experience.

Step 4: Add Default Blockify Datasets

Retrieval-Augmented Generation (RAG) is powered by datasets. Pre-loading sample or custom Blockify datasets enables users to immediately experience AirgapAI's enhanced accuracy with enterprise data.

1. Identify Dataset Storage Location: AirgapAI stores datasets in another specific AppData location: C:\Users\[Username]\AppData\Roaming\airgap-ai-chat\CorpusRepo
2. Copy Sample or Custom Datasets:
   • Navigate to the /datasets/ directory within your extracted installer folder.
   • Select the desired sample datasets (e.g., CIA_World_Factbook_US.jsonl) or any custom Blockify datasets you have prepared.
   • Copy these .jsonl files into the CorpusRepo directory.
3. Golden Image Integration: Similar to models, configure your imaging solution to copy these default datasets to each new user's AppData\Roaming\airgap-ai-chat\CorpusRepo upon first login or as part of the Default User profile. Note: For optimal accuracy (78 times improvement), larger corpora should be converted to Blockify format. Local on-device Blockify will be available in the beginning of the third quarter of 2025.

Step 5: Configure First-Launch Wizard Suppression and Default Settings

Pre-loading models and datasets will largely suppress the first-launch onboarding wizard. However, information technology teams may want to set additional default configurations.

1. Default Chat Style and Profile Name: While not explicitly detailed as a command-line option, the documentation shows the onboarding wizard prompts for a display name and chat style. When pre-loading models and datasets, the application often defaults to standard settings. If customization is required, Iternal Technologies can provide guidance on configuration files to modify within the application's installation directory or user profile that dictate these settings.
2. Initial Model Benchmarking: On first model launch, AirgapAI offers to benchmark hardware. You can configure a policy or script to run this benchmark automatically or suppress it, depending on your deployment strategy. The documentation states that context-size limits remain at a conservative 2,000 tokens until a benchmark is completed, so it is recommended to run it.

Step 6: Optional: Dell Technologies Dell Pro Artificial Intelligence Studio Support

For organizations leveraging Dell Technologies' Dell Pro Artificial Intelligence Studio (DPAIS), AirgapAI Chat supports native integration.

1. Install DPAIS Files: As the Information Technology Systems administrator, install the required files to enable a large language model via DPAIS (both Intel and Qualcomm are supported).
2. Validate DPAIS Services: Ensure DPAIS services are running and validate that the local large language model Application Programming Interface endpoints can be called.
3. Set Environment Variable: Open Powershell and input the following command:

   [System.Environment]::SetEnvironmentVariable("DPAIS_ENDPOINT", "http://localhost:8553/v1/openai", "User")
   

4. Relaunch AirgapAI Chat: After setting the environment variable, relaunch the AirgapAI Chat application. The DPAIS large language models available will automatically appear in the model selection menu in the settings page.

Key Enterprise Deployment Features and Management

Beyond initial installation, AirgapAI provides robust features and management capabilities for information technology teams.

Blockify Technology for Unparalleled Data Accuracy

Blockify is the ultimate data management solution for large language models at scale, allowing you to create a single source of truth optimized for precise answers.

• Ingestion Process: Blockify ingests large datasets (e.g., thousands of sales documents or Request for Proposal responses) and condenses them into concise, modular "blocks" of data. Each block includes a name, a critical question, and a trusted answer.
• Security and Data Lifecycle Management: Each block is tagged with rich metadata, including classification, permissions, and classification levels, to support zero-trust environments.
• Accuracy Improvement: This process can reduce original data size by up to 97.5 percent and improve large language model accuracy by 7,800 percent (78 times). This transforms enterprise data into a "trusted corpus of information" for your "confidential artificial intelligence chat application."

Role-Based Workflows and Entourage Mode

AirgapAI offers customizable features to enhance user productivity and decision-making.

• Role-Based Workflows: Quick Start workflows can be tailored for different roles (e.g., procurement, legal, engineering), pre-configuring prompts that automatically select relevant curated datasets. This supports a "build your own artificial intelligence assistant" approach for specific departmental needs.
• Entourage Mode: A unique feature allowing users to interact with multiple artificial intelligence personas simultaneously. For example, a "CIA analyst" persona and a "military tactician" persona can provide distinct answers to the same question, offering multi-perspective views for high-stakes decision-making. This is ideal for "customizable artificial intelligence personalities" in complex scenarios.

Ongoing Updates and Maintenance for Information Technology Teams

Managing updates is critical for enterprise applications. AirgapAI streamlines this process.

• Synchronized Update Cadence: AirgapAI's update cadence can be synchronized with your typical operating system or enterprise software update cycle. Information technology can deploy new versions and dataset updates through familiar image management solutions.
• Configuring Update Sources: The updaterConfig.json file located at C:\Users\John\AppData\Local\Programs\AirgapAI Chat\resources\auto-updater\ allows you to change the file server update location. This enables information technology to host updates on a local server for maximum security and control, ensuring "local secure artificial intelligence software" updates without external network dependencies. Example updaterConfig.json structure for local update server:

  {"win32-x64-prod":{"readme":"","update":"http://yourlocalserver.com/releases/prod/public/chat-assistant/prod/public/1.0.2/AirgapAI Chat Setup 1.0.2.exe","install":"http://yourlocalserver.com/releases/prod/public/chat-assistant/prod/public/1.0.2/AirgapAI Chat Setup 1.0.2.exe","version":"1.0.2"}}
  

  Replace http://yourlocalserver.com with the Uniform Resource Locator of your internal update server.

Security Review and Validation Checklist for Secure Install

Deploying AirgapAI via enterprise imaging enhances your security posture. Use this checklist to validate your secure installation:

• Local Data Storage: Confirm that all artificial intelligence models, embeddings, and Blockify datasets are stored exclusively on the local device within user profiles, not on external network shares or cloud storage. This ensures a "truly private artificial intelligence assistant."
• No Network Ingress/Egress for Artificial Intelligence Operations: Verify that AirgapAI's core large language model inferencing and Retrieval-Augmented Generation operations do not initiate outbound network connections or require inbound access. This confirms its "air-gapped" nature.
• Integration with Zero-Trust Frameworks: Ensure Blockify's metadata tagging for classification and permissions aligns with your organization's zero-trust security policies.
• Controlled Update Mechanisms: Validate that application and dataset updates are managed exclusively through your chosen information technology deployment tools (e.g., Microsoft Intune, local file server) and not through public cloud channels unless explicitly configured.
• User Profile Isolation: Confirm that each user leveraging AirgapAI accesses personalized, role-specific datasets stored securely within their isolated user folder, preventing cross-user data exposure.
• Operating System Security: Verify that the artificial intelligence personal computer adheres to all standard operating system security hardening guidelines, including firewalls, antivirus, and encryption.

Benefits for Your Organization: A Secure, Cost-Effective Artificial Intelligence Future

By integrating AirgapAI into your enterprise imaging process, your organization achieves immediate and long-term benefits:

• Unparalleled Security and Data Privacy: Keep your artificial intelligence conversations private and confidential, eliminating concerns about data leakage or compliance breaches associated with cloud-based solutions. This offers a "privacy first artificial intelligence assistant" and is perfect for "artificial intelligence for confidential chats."
• Significant Cost Savings: Move away from expensive per-user, per-month subscriptions to a one-time perpetual license. This offers a "no subscription artificial intelligence application" that is "one device artificial intelligence license" and is considerably cheaper than alternatives.
• Trusted and Accurate Insights: Leverage Blockify technology to empower your employees with 78 times more accurate artificial intelligence responses, reducing validation time and increasing confidence in artificial intelligence-generated content.
• Empowered and Productive Workforce: Provide your teams with a cutting-edge generative artificial intelligence chat assistant that is easy to use, works offline, and integrates seamlessly into their daily workflows, whether in the office or in disconnected environments.
• Simplified Information Technology Management: Streamline deployment, updates, and governance through existing enterprise imaging and management tools, freeing up valuable information technology resources.
• Future-Proof Artificial Intelligence: With "local model support" and "bring your own model artificial intelligence" capabilities, AirgapAI ensures flexibility to adapt to evolving artificial intelligence needs.

Training and Support

Iternal Technologies is committed to your success. We offer:

• Comprehensive Documentation: Access our online enablement page which includes step-by-step videos, Frequently Asked Questions, user guides, and troubleshooting tips.
• Personalized Training: We offer personalized training sessions as an add-on service, ensuring your team can fully leverage AirgapAI's capabilities.
• Dedicated Support: Our customer success team is available for follow-up calls, additional workshops, and technical assistance. For additional questions, contact the product team at support@iternal.ai.

Empower your workforce with a secure, efficient, and cost-effective artificial intelligence solution that puts data privacy first.

Download the free trial of AirgapAI today at: https://iternal.ai/airgapai