genAI Capabilities Canvas - a visual tool (obviously 😉) to help you plan your generative AI app
The genAI Capabilities Canvas is a visual thinking tool designed to help you understand and plan for the capabilities of generative AI in your application.
The genAI Capabilities Canvas is structured around five key areas, each offering an orthogonal perspective on generative AI. This multi-faceted approach ensures that you consider all essential aspects of AI capabilities:
1. Modalities
Focuses on the types of data your AI will interact with, such as text, voice, or images. It helps you consider the scale of the output in relation to the input.
2. Context
Explores how the AI maintains a dialogue, understands the scope of the conversation, answers queries responsibly, and performs semantic searches.Â
3. Knowledge
Examines the AI's foundational understanding, including common sense, world knowledge, and the ability to incorporate your own content and specialized training.
4. Agency and Autonomy
Looks at the AI's decision-making capabilities, its limitations in computational tasks, and its ability to plan and execute tasks independently.
5. Styling
Considers how you can adjust the tone, style, and even the content specificity of the AI's output through fine-tuning.
These areas are designed to be complementary, providing a multi-faceted, 360-degree view that aids in both strategic planning and effective implementation.
template:
https://app.mural.co/template/8293d773-8b1a-4585-8ac2-c48b34e484cd/8e99b7aa-d89c-4950-add1-42789e0aa587
learn more:
https://genaiarchitecture.com/plans/336881?bundle_token=6cb1b6f1a403d4e09c9f119193aae0e7&utm_source=manual