INSYTE Framework Worksheet
A step-by-step guide to evaluate your intelligent system
INSYTE Framework - Worksheet and Radar Chart Generator
Welcome to the Framework of Intelligent Systems ('INSYTE') for the classification of AI-enabled systems.
The INSYTE framework considers the characteristics of intelligent systems along eight key dimensions, each of which may be instantiated at a level between 0 and 5. This worksheet will help you determine and select your system’s level on each of INSYTE’s eight dimensions. Each dimension is then represented as an axis on a radar chart. At the end of working through this worksheet, you will be able to generate and download your system's intelligence "pattern" on the INSYTE radar chart.
Important Information
Browser Compatibility: This tool has been primarily tested on Google Chrome. Users may experience issues with chart export and other functionality when using Safari or other browsers.
Session Data: Your assessment data is stored only in your browser's local storage. You cannot save intermediate results between sessions. Please complete your system assessment in one session without closing the browser tab.
Citation
Zoe Porter, Radu Calinescu, Ernest Lim, et al. (2025). INSYTE: A Classification Framework for Traditional to Agentic AI Systems. ACM Transactions on Autonomous and Adaptive Systems. (August 2025).https://doi.org/10.1145/3760424
The eight INSYTE dimensions are organised into four categories as follows :
| Category | Dimension | Description |
|---|---|---|
| System Design | 1. Underspecification | Ability of the system to accomplish its objectives without an explicit specification of how to do so |
| 2. Adaptiveness | Ability of the system to accomplish its objectives under operational uncertainty and change | |
| System Functionality | 3. Breadth of functionality | Ability of the system to perform a range of different task types or generate a range of output types |
| 4. Depth of functionality | Ability of the system to execute complex tasks, or to produce sophisticated outputs | |
| Operating Environment | 5. Environmental diversity | Ability of the system to accomplish its objectives in a diverse, open-world environment |
| 6. Environmental dynamism | Ability of the system to accomplish its objectives in a frequently and rapidly changing environment | |
| Operational Independence | 7. Operational independence from intervention | Ability of the system to accomplish its objectives without human intervention |
| 8. Operational independence from oversight | Ability of the system to accomplish its objectives without human monitoring |
The process
The INSYTE framework can be applied to any AI-enabled system, including decision-support systems such as deep-learning diagnostic tools, and cyber-physical systems such as autonomous robots. You will now go through stages of applying the framework to your AI-enabled system.
Ready to begin?
Click the "Next" button below to start the assessment process. You'll begin by selecting how many system variants you want to evaluate.