Intro
In generative AI, prompts are essential for guiding the AI's output. Clear and effective instructions ensure that the output aligns with the user’s needs.
Quick prompts are predefined actions integrated into workflows for easy access. These types of prompts are crafted by experts known as prompt engineers, who focus on the efficiency of AI.
[internal_only]For more information on all types of prompt patterns, see Designing Effective AI Prompts.[/internal_only]
Content generation with quick prompts
When to Use
Use quick prompts when:
- Tasks are repetitive or for common actions within a workflow.
- The system can only assist with specific actions.
- Users lack expertise in the subject matter.
- It’s crucial to minimize any bias introduced by users’ writing prompts.
- Maintaining consistent and predictable output is essential.
Don’t use quick prompts:
- When a user’s intent is unpredictable.
- When users need more flexibility in directing the output of the AI model.
- For non-AI functions.
- For multiple fields in the form – use actions at the form level instead.
- When implemented through the AI button or menu button alongside the AI writing assistant.
- Use quick prompts with the AI notice base concept to inform users they’re interacting with AI.
- Clearly identify quick prompts with the AI icon.
- Make sure that the AI output is unbiased, inclusive, and aligns with the company tone and identity.
- Use recommended AI action labels.
- Follow the action placement guidelines and use the emphasized button style only when a quick prompt is the primary action for the page.
- Ensure users can complete the task without using quick prompts and without using AI.
- Clearly identify quick prompts with the AI icon.
- Make sure that the AI output is unbiased, inclusive, and aligns with the company tone and identity.
- Use recommended AI action labels.
- Follow the action placement guidelines and use the emphasized button style only when a quick prompt is the page’s primary action.
- Ensure users can complete the task without using quick prompts and without using AI.
Components
The quick prompts pattern extends the following components to enable new AI-specific interactions:
AI button
AI button with a quick prompt
AI menu button
AI menu button with multiple quick prompts
For more information, see [internal_only]AI Icon, [/internal_only]Menu Button, Menu, and the Terminology section.
Behavior and Interaction
Generating content
Starting the generation process
Generating content with quick prompts
During generation
During content generation with quick prompts
Refining
When generation finishes or is stopped, the button transitions to an AI menu button, offering quick prompts users can apply to refine the output. Users can also edit the content directly.
Refining generated content with quick prompts
For information on handling warnings when versioning is not supported, see AI Writing Assistant and Regenerate.
For information on how to display the text area during generation, see Busy Indicator.
Handling Errors
If the quick prompt is interrupted or fails, follow the guidance for error messages in the message handling article.
For a consistent user experience, we suggest using the following error message:
Something went wrong while generating your content. Please try again.
Error handling for content generation using quick prompts
Grouping
AI menu grouping quick prompts in a submenu
Grouping AI and non-AI actions
Group related actions by purpose or outcome, whether they are AI or non-AI functions. When combining both in the same menu, use the AI icon to identify AI-powered quick prompts and visual dividers to clearly separate them from non-AI actions.
For more information and labels for submenus, refer to the terminology section.
AI menu grouping non-AI actions and quick prompts in a submenu
Terminology
Use the default labels provided below for text generation scenarios. To ensure a consistent and familiar experience for users, we recommend using the suggested wording. Only make changes to the default text if absolutely necessary for your specific use case.
Recommended AI UI text for AI text generation and transformation
For actions not covered above, apply the following guidelines:
- Use a verb in the imperative.
- Keep AI action labels as short as possible while prioritizing clarity for users.
- Use the same AI action labels consistently.
For more information, see UI Text Guidelines for SAP Fiori Apps.
Responsible AI
Appropriate use of quick prompts
Quick prompts are designed to make interactions easier by providing clear, predefined options. They help users complete tasks faster and reduce confusion. However, it's important to use them appropriately. Quick prompts should align with the user's needs. In complex scenarios where more detail or context is required, they may fall short. In such cases, guided or custom prompts can lead to more accurate and comprehensive responses.
[internal_only]Additionally, ensure that using quick prompts aligns with SAP’s commitment to responsible AI practices and adheres to our SAP Global AI Ethics Policy by completing the AI Ethics Assessment Process. For use cases classified as standard, we recommend completing the AI Ethics Policy Self-Assessment.[/internal_only]
User autonomy
Empower users by enabling them to disable quick prompts and AI features in their workflow. Ensure alternative methods are available for users to complete their tasks without relying on AI.
AI transparency
Ensure clarity in the deployment of quick prompts, making their use and purpose transparent to users. Use clear indicators, like AI icons or buttons, to ensure users know when they’re interacting with AI. Follow our design guidelines to maintain consistency in how these features are presented.
Prevent bias in prompt design
Ensure the prompt design process actively involves measures to prevent bias. Like biased training data, poorly designed prompts can produce skewed or harmful results.
Advocate for a diverse group of users and technical experts to focus on thoughtful prompt design, conduct comprehensive evaluations of user needs and outcomes, and iterate to ensure that AI-generated content is as free from bias as possible.
Prompt Engineering
Prompt engineering is the process of designing and refining instructions to guide the behavior and output of generative AI models.
Your product team is responsible for engineering effective prompts tailored to the underlying AI model, ensuring users get the desired results when using quick prompts.
Follow the [internal_only]best practices[/internal_only][external_only]best practices[/external_only] and, where necessary, use advanced LLM techniques like embeddings and fine-tuning to get the best outcomes.
Helpful Terms
Foundation models
Large deep learning models trained on massive unlabeled data with self-supervised learning. They provide general-purpose capabilities across domains (text, images, audio, multimodal).
Large language models (LLMs)
A subset of foundation models specialized in natural language. Trained on vast text datasets, they excel at understanding and generating human language and can be fine-tuned for specific tasks or domains.
Prompt engineering
The process of designing and refining instructions to guide the behavior and output of generative AI models.
Embeddings
A way to represent language or data as numbers that capture meaning and similarity. This enables AI to recognize when different words or phrases convey similar ideas, allowing it to retrieve relevant information even without exact word matches and to generalize beyond its training data.
Fine-tuning
Fine-tuning LLMs is the resource-intensive process of customizing a pre-trained language model on specific tasks or datasets to make it more proficient and accurate in generating relevant text.
Related Links
Guidelines
Implementation
- Button (SAPUI5 samples)
- AI Button (UI5 Web Components documentation)
- Quick Prompts (UI5 Web Components documentation)
Implementation
- Button(UI5 Web Components documentation)
- AI Button (UI5 Web Components documentation)
- Quick Prompts (UI5 Web Components documentation)