Generative AI
Generative AI is a category of artificial intelligence that creates new content, such as text, images, audio, or even video, based on the data it has been trained on. It works by using models, such as neural networks, to generate outputs that resemble the training data in style or content but are new and unique.
Generative AI is a rapidly evolving field that is transforming how we create and interact with digital content. Its applications span numerous industries, including education, where it has the potential to transform how we teach and learn.
Protect yourself when using generative AI! Be careful not to share sensitive information, such as university business, personal details, proprietary research, or any data protected by law, when using these tools. For information on responsible use, please refer to our Guidelines for AI use at KSU.
Artificial Intelligence at KSU—Return to landing page
What Can Generative AI Do?
Generative AI can create:
- Text: Generate, outline, draft, and refine emails, essays, and articles; answer questions; organize complex information; and analyze documents.
- Images: Create original artwork, enhance existing images, and generate realistic faces or objects.
- Music & Audio: Compose music and produce human-like speech or sound effects.
- Video: Generate animations, apply visual effects, and create high-quality video content from minimaHow to Create Effective Prompts l input.
How to Use Generative AI Tools
Clearly define your objective, then provide detailed prompts to guide the AI. Always review and refine the output to ensure accuracy and quality.
Generate Text
Generative AI can create all kinds of text, including emails, essays, stories, poems, social media posts, speeches, interview questions, and more! Going a step further, generative AI can also revise text, whether it’s text you wrote or text the AI itself generated.
A note on AI output: AI chatbots generate text, not necessarily facts. Be careful to check and double-check information generated by AI.
Effective Prompts for Generating Text
When it comes to creating effective prompts, it’s all about giving clear instructions and engaging with AI the way you would with a fellow human! Here are additional tips to help you craft prompts that unlock the full potential of Copilot:
- Consider your desired outcome: No matter the type of content you want Copilot to generate, start by asking yourself, “What result do I hope to achieve?”
- Structure your prompt with purpose: Place the most important information at the beginning of your prompt.
- Use natural language: Communicate with Copilot just as you would with a human.
- Provide context: Share background information as well as details about the intended audience.
- Define the task: Try using these words to direct the AI: explain, describe, translate, create, write, predict, advise, imagine, summarize, revise.
- Be specific: The more precisely worded your prompt, the more precise the generated content will be.
Examples of Effective Prompts
Example 1
Write a motivational speech for a group of researchers facing tight deadlines, highlighting the importance of teamwork, resilience, and innovation.
This prompt is specific, clearly identifying the audience (IT professionals) and the context (tight deadlines). It outlines key themes (teamwork, resilience, innovation) that guide the AI's focus, ensuring relevant content.
Example 2
Create a blog post about the benefits and risks of using generative AI to generate lesson plans. Include examples of how AI can help instructors improve students' learning experiences. Note the security risks and provide instruction on how to use AI safely.
This prompt specifies the genre (a blog post), topic (using AI to generate lesson plans), and desired content (examples of ways AI can help, plus information about using AI safely), making it easier for the AI to generate focused and informative content.
Example 3
Describe a day in the life of a phlebotomist at a busy hospital in a big city, focusing on key tasks, challenges, and the skills required to succeed.
This prompt sets a clear scenario (day in the life) and focuses on specific aspects (tasks, challenges, skills), prompting the AI to create a narrative that offers insight into the role.
Generate Images
Unleash your inner artist with generative AI! Many chatbots, including Copilot, can create graphics, illustrations, and even photorealistic images.
Effective Prompts for Generating Images
Follow these suggestions to endure your prompts help generative AI create just the graphic you imagine:
- Color code: Describe the exact hues using color codes (e.g., #FFC629 for Hooty Hoo Gold) or evoke emotions with descriptive terms (e.g., a sunset's fiery orange or the calming blue of a twilight sky).
- Shift your perspective: Imagine the scene. Are you looking up at a towering skyscraper (low-angle) or down a whimsical path disappearing into a forest (bird's-eye view)? Consider including this information in your prompts.
- Provide inspiration: Channel your artistic idols! Go for a classic Renaissance portrait (think Da Vinci's Mona Lisa) or a swirling, expressive scene reminiscent of Van Gogh's “Starry Night.”
Examples of Effective Prompts
Example 1
Create an illustration of a serene landscape with a sunset over a mountain range with a calm lake in the foreground and a few birds flying in the sky.
This prompt describes the elements to be included in the illustration, including elements in the foreground and larger image, and words like "serene" and "calm" help the AI understand the desired mood and atmosphere.
Example 2
Design an abstract image with vibrant colors, geometric shapes, and a sense of movement and energy.
This prompt clearly outlines the style, key elements, and visual details to be included in the image, such as vibrant colors, geometric shapes, and a sense of movement and energy. The phrase "sense of movement and energy" sets the tone for the image.
Example 3
Depict a scene from the Renaissance era, with artists painting, people in period clothing, and classical architecture in the background.
This prompt is clear and specific: It outlines the key elements to be included in the scene, such as artists painting, people in period clothing, and classical architecture. It provides rich visual details, including specific elements like "artists painting" and "classical architecture." The prompt also sets a historical context, guiding the AI to create an image that reflects the artistic and cultural characteristics of the Renaissance period.
Want to learn more? Check out these resources:
