Nexthink Adopt Guide Recorder - AI model card

Model details

Description

The Nexthink Adopt Guide Recorder feature generates and suggests guide content based on user input from the Adopt Editor—using the Nexthink Extension—and stored guide content. By leveraging LLM models hosted on AWS Bedrock within the Nexthink Infinity infrastructure, this feature automatically develops or suggests guide content by interpreting recorded user in-app actions or existing guide walkthroughs.

The capability is embedded in the guide creation and management process within Adopt Editor. These AI models process stored guide content or recorded steps to generate content for guides—ensuring quality, consistency, time-effective development, and accessibility to user personas that develop guides without necessarily being technically proficient using Adopt Editor.

Refer to the FAQ documentation for more information about Nexthink, and the Recording Walkthrough guides documentation for details on developing Adopt guide content with AI.

Inputs and outputs

Inputs

• In-app sequence step recordings (events) done by Admins using Adopt Editor from the Nexthink Extension, when developing new guide content.

• Existing guide content created and edited by Admins using Adopt Editor from the Nexthink Extension, including guide name, step content (HTML), and step names, configured in the Nexthink Infinity UI.

Outputs

• Proposed listed steps in Adopt Editor, which Admins can accept, edit, or reject to record in-app sequence steps again.

Intended use

Primary intended users

Adopt administrators responsible for creating and managing guides in Nexthink.

Subject matter experts (SMEs) responsible for developing/reviewing guide walkthroughs and in-app instructional content using Adopt Editor.

Out-of-scope use cases

The Nexthink Adopt Guide Recorder for AI content generation is solely used for developing Adopt walkthrough step-by-step guides or getting content suggestions based on user-recorded actions and stored guide content in Nexthink Infinity. It is not intended to process sensitive information and does not have access to web application data.

Model data

Data flow

Nexthink Adopt Guide Recorder feature integrates AI to enhance content generation by utilizing the Nexthink Extension to process stored guide content in Nexthink Infinity, as well as in-app user actions and metadata, to suggest and generate content.

This process leverages AWS Bedrock for input interpretation and output.

1. Browser - User in Adopt Editor from the web application: Starts and stops recording of in-app actions in the Adopt Editor—Nexthink Extension. Sends interaction data (clicks, keystrokes, element selectors) and recording metadata (timestamps, session IDs) to the extension service.

2. Nexthink Extension (UI Events Collector): Captures in-app user events asynchronously. Bundles data and forwards it to the Nexthink Adopt backend.

3. Nexthink Adopt backend: Acts as intermediary, formatting and validating recorded events. Prepares data for AI processing without directly storing user-sensitive information.

4. AWS Bedrock: Processes the structured event data to generate draft guide steps. Predicts next steps, applies formatting or localization rules, and returns a draft guide. All data processing stays entirely within the Nexthink AWS environment.

5. Nexthink Extension (Review of LLM proposal): Receives AI-generated guide steps and presents them to the user. The user can accept, adjust, or reject any step. Adjustments are sent back to storage as the finalized guide.

6. Nexthink Infinity: Only stores the finalized guide content after user review. Nothing else is stored, not even temporarily.

7. Nexthink Extension (Guide in Adopt editor): AI-generated and confirmed AI content is listed in Adopt Editor, step by step.

Evaluation data

Nexthink employs validation and monitoring practices to ensure that AI-generated content remains accurate and reliable. Continuous oversight helps proactively detect issues such as formatting errors or inconsistencies in AI-generated guide steps. The Nexthink Adopt Guide Recorder for AI content generation enables users to review, adjust, or reject generated content directly within Adopt Editor—in the Nexthink Extension—ensuring clarity and correctness before guides are published. This human validation complements AI-generated output, maintaining translation consistency and improving the overall employee experience over time.

AI is getting better every day, but it can still make mistakes. Nexthink recommends that Admins always review AI-generated translations before publication to ensure accuracy and clarity.

Training data

Nexthink Adopt Guide Recorder uses off-the-shelf LLMs hosted on AWS Bedrock. Nexthink does not fine-tune this model and does not use Customer Data to train AI models.

Preprocessing data

Guide content created in the Nexthink Extension (guide name, step content in HTML, and step names) is retrieved from the Nexthink Data Store, combined with the recorded in-app user actions (input events: clicks, keystrokes, element selectors, and metadata) from Nexthink Extension. This structured package of guide content and recorded in-app user actions is then sent to the AWS Bedrock hosted AI model.

Implementation information

Hardware

Models run within AWS infrastructure in the customer’s geographic region using AWS Bedrock.

Software

The Nexthink Adopt AI for content development uses an off-the-shelf AI LLM model hosted on AWS Bedrock. Nexthink does not use Customer Data to train its AI models.

Security

Nexthink employs HTTPS and AES-256 encryption to secure data both in transit and at rest. Nexthink's use of standard encryption methods aligns with industry best practices to prevent unauthorized access and protect data processed by AI features. Visit Nexthink Security Portal to learn more about Nexthink's commitment to information security.

Caveats and recommendations

Risk management

Risk	Mitigation
Hallucination and bias propagation	Model hallucinations and biases are mitigated by Nexthink through continuous performance monitoring and regular model updates. For generated guide content, administrators can review and adjust guide steps to ensure accuracy.
Inaccuracy of outputs	The Adopt Guide Recorder capability can occasionally produce inaccurate or inconsistent translations, as AI systems continue to evolve. Admins are therefore encouraged to review and adjust generated content before publishing to ensure clarity, accuracy, and consistency with organizational terminology. While AI accelerates the translation process, human oversight remains essential to mitigate errors and maintain quality.

Ethical considerations

Nexthink adheres to both national and international AI guidelines and best practices, emphasizing the responsible and ethical development of AI. In compliance with the EU AI Act, Nexthink has developed a comprehensive AI compliance framework. Each AI component is reviewed by a dedicated AI Compliance Team comprising Legal, Privacy, and Security experts, among others.

AI Limitations

While the Adopt Guide Recorder capability can be highly beneficial in accelerating the development of guide content, it is important to recognize its current limitations. AI models are still evolving and may occasionally produce errors, inconsistencies, or outputs that deviate from expected results.

To mitigate these risks, Customers may:

• Cross-Check AI outputs: Validate AI-generated results against reliable sources and internal benchmarks.

• Implement human oversight: Use AI as a supporting tool rather than a decision-making authority, ensuring that critical guide outputs are reviewed by qualified individuals.

• Provide feedback: When inaccuracies are identified, share them with the AI provider (if applicable) to contribute to model improvement.

Customers should clearly communicate to employees the limitations of Nexthink Adopt AI for the content generator and how to best adhere to the risk-management and mitigation practices outlined above.

FAQ

How does Nexthink Adopt Guide Recorder leverage Artificial Intelligence?

The Adopt Guide Recorder feature leverages an AI model hosted on AWS Bedrock to automatically generate Nexthink Adopt guide content into sequenced steps. The model processes in-app actions data recorded in Adopt Editor—Nexthink Extension. By applying natural language processing and logic, the AI system generates guide steps that reflect the recorded structured sequence in the web application UI.

This use of AI reduces the need for manual guide step development, including content writing, element selector allocation, and key page mapping. This AI feature ensures scalability across multiple departments, allowing Guide Admins to deliver contextual guidance globally with greater efficiency.

Does Nexthink Adopt Guide Recorder make any automated decisions?

The Adopt Guide Recorder for AI content generates structured guide steps, but administrators retain full control over reviewing, adjusting, and publishing the Adopt guides. This design ensures that admins maintain oversight and decision-making authority, with AI functioning as a support tool to accelerate development rather than an autonomous system.

How can I identify if I'm developing guide content with AI?

When using recording guide steps from the Adopt Editor—Nexthink Extension—the system displays the ✦ sparkles icon to indicate AI-generated content or insights.

Does AWS Bedrock process Personal Data and Personal Identifying Information (PII)?

Nexthink is fully committed to protecting its Customers' data. When using AI functionalities hosted on AWS, all processing takes place within the AWS region aligned with the Customer's Nexthink deployment. At no point is Customer Data or Personal Data shared with or hosted by the AI tool providers themselves. Neither Nexthink nor AWS uses Customer Data or Personal Data for model training purposes. Additionally, any Customer Data processed through AI functionalities, including any Personal Data or PII that may appear in Guides, is subject to the same retention periods and protective measures as all other Customer Data within the Nexthink solution.

Can AWS Bedrock see the guide recordings from users?

AWS Bedrock has no access to or any visibility of user screens, as Nexthink Extension shares recorded in-app user actions and related metadata (timestamps). No image or video is captured, stored, or shared.

Can AWS Bedrock leverage user data to train its models?

No, AWS Bedrock cannot and does not use data submitted by customers via its APIs to train or by the large language model (LLM) provider within AWS Bedrock to improve its models.

Where does AWS Bedrock process its data?

All data processing stays entirely within the Nexthink AWS environment.

How does Nexthink ensure user training?

The AAI Adopt content generator feature is designed to be easily accessible from Adopt Editor. That said, Nexthink provides both documentation and a video course about the Nexthink Adopt through its Documentation and Learn portals.

How does Nexthink inform users about the changes to this Guide Recorder AI feature for Adopt?

Nexthink provides documentation and training resources about Adopt, including how to use the AI content generation capability, or changes that may affect user experience through 'What’s New' notifications, email communications, and Documentation updates.

Finally, customers are promptly informed via an email communication in case of the planned introduction of a new sub-processor used by AI features.