Ai for Creative Projects

• A. A system that only retrieves and copies existing web pages
• B. A system that learns patterns from data and produces new content such as text, images, or music ✓
• C. A system that can only perform arithmetic calculations
• D. A system that requires a human to write every output manually

Why: Generative AI models learn patterns from large datasets and use them to create new content, rather than simply copying or retrieving existing material.

• A. A hardware component that stores training data
• B. A bug that causes the model to crash
• C. The input text or instructions given to an AI model to guide its output ✓
• D. The final file format of a generated image

Why: A prompt is the input, usually written instructions or descriptions, that tells the AI model what kind of output to produce.

• A. AI makes human creativity unnecessary
• B. AI can only imitate creativity and never contribute ideas
• C. AI can serve as a collaborator that expands ideas, while human judgment guides taste, intent, and final choices ✓
• D. Creativity is purely mechanical and AI fully replicates it

Why: Most creative practitioners view AI as a collaborative tool that generates options and sparks ideas, while humans provide direction, evaluation, and meaning.

• A. It is always identical to existing works
• B. It can produce outputs that are inaccurate, generic, or inconsistent, requiring human review ✓
• C. It cannot generate any text at all
• D. It can only produce audio files

Why: AI outputs can be plausible but wrong, overly generic, or inconsistent, so human review and editing remain essential for quality.

• A. Ethics only matters for scientific research, not art
• B. AI tools never raise questions about attribution, data sources, or originality
• C. Issues such as credit, consent for training data, and transparency about AI use affect fairness and trust ✓
• D. Ethical considerations prevent AI from running faster

Why: Creators must think about attribution, how training data was sourced, and whether they disclose AI involvement, since these affect fairness, trust, and legal considerations.

• A. Decision tree classifiers
• B. Diffusion models ✓
• C. Linear regression models
• D. K-means clustering

Why: Diffusion models learn to progressively remove noise from random data, guided by a text prompt, to produce images. Popular image generators such as Stable Diffusion and DALL-E use this approach, while the other options are not designed for image generation.

• A. To store the final image file on the cloud
• B. To describe the desired output so the AI can generate matching content ✓
• C. To change the resolution of an image after it is created
• D. To install new software plugins

Why: A prompt is a natural language description that tells the AI what to create, including subject, style, lighting, and composition. The quality and specificity of the prompt strongly influence the result.

• A. Accept the first output without changes
• B. Iterate on the prompt by adding specific details about style, mood, or color ✓
• C. Avoid using any descriptive words in the prompt
• D. Only use images created by other people

Why: Iterating on prompts with more precise details about style, mood, lighting, or color typically produces results closer to the creator's vision. Treating generation as a process of refinement is a core skill in AI-assisted design.

• A. Convert a photo into a vector graphic
• B. Regenerate or replace a selected area of an existing image while keeping the rest intact ✓
• C. Automatically print the image on fabric
• D. Compress the file size without losing quality

Why: Inpainting lets you mask a portion of an image and have the AI fill that region based on a prompt and the surrounding context. It is useful for fixing details, removing objects, or adding elements without redoing the whole composition.

• A. AI tools never produce any usable images
• B. Licensing terms and the origin of training data can affect how outputs may legally be used ✓
• C. Copyright only applies to hand-drawn art
• D. Attribution is required only for photographs

Why: Different AI platforms have different terms of service, and questions about training data and ownership of generated works are still evolving. Creators should check licensing terms and disclose AI involvement where required to use outputs responsibly.

• A. To convert images into written descriptions
• B. To create new images from natural language prompts ✓
• C. To edit the colors of existing photographs
• D. To compress large image files into smaller ones

Why: Text-to-image models take a written prompt describing a scene or concept and generate a new image that matches that description. Converting images to text is captioning, editing is a separate task, and compression is unrelated to generation.

• A. The file format used to save the final artwork
• B. The hardware used to run the AI model
• C. The text instructions given to the AI to guide the output ✓
• D. The licensing agreement for the generated image

Why: A prompt is the text input that tells the AI what to create, including subject, style, lighting, and composition. The quality of the prompt strongly influences the quality and relevance of the generated image.

• A. Using random, unrelated keywords in each prompt
• B. Specifying an art style, medium, or artist reference in the prompt ✓
• C. Generating images at the lowest possible resolution
• D. Avoiding any descriptive language

Why: Explicitly naming a style, medium (such as watercolor or oil painting), or lighting approach gives the model clearer guidance, producing more consistent results across multiple generations.

• A. A dataset of training images
• B. A number that initializes the random noise the model starts from, enabling reproducible results ✓
• C. The final approved version of an artwork
• D. A type of neural network layer

Why: The seed determines the starting random noise pattern. Using the same seed with the same prompt and settings can reproduce a similar image, which is useful for iterating on a design without losing what you liked.

• A. AI-generated images are always in the public domain with no restrictions
• B. Copyright and licensing rules for AI outputs vary by jurisdiction and platform, so creators should check terms and disclose AI use where appropriate ✓
• C. Attribution is only required for hand-drawn art
• D. Copyright law never applies to digital media

Why: Legal status of AI-generated works differs across countries and platforms, and training data and usage terms can affect what you may do with outputs. Checking terms and being transparent about AI involvement are responsible practices for creative work.

• A. By copying and pasting existing songs from a database without modification
• B. By learning patterns in existing audio or symbolic music data and generating new sequences that follow those patterns
• C. By randomly selecting notes from a fixed list of pitches with no learned structure
• D. By requiring a human to play every note manually before the AI can process it

• A. Raw photographs of musical instruments
• B. MIDI files, which encode notes, timing, and velocity as data
• C. Plain text transcripts of song lyrics only
• D. Video recordings of live concerts without audio

• A. The overall loudness of a sound over time
• B. The tempo of a musical piece measured in beats per minute
• C. The tonal color or quality that distinguishes one sound from another at the same pitch and volume
• D. The length of silence between two musical phrases

• A. Style transfer using neural networks
• B. Simple volume normalization
• C. Manual equalization with a graphic EQ
• D. Reversing the waveform of the original audio

• A. Ensuring the AI output is always in a major key
• B. Checking the licensing and copyright status of training data and the generated output
• C. Using only acoustic instruments in every composition
• D. Limiting the song length to under three minutes

• A. Asking the AI to write the entire novel without any input from you
• B. Using the AI to brainstorm multiple premises, then choosing and refining one yourself ✓
• C. Copying the first output the AI produces without reviewing it
• D. Avoiding AI entirely because it cannot be creative

Why: AI works well as a brainstorming partner that generates many options quickly. The writer's judgment is still needed to select, shape, and refine the idea into something original and personal.

• A. AI cannot produce any dialogue at all
• B. AI output can feel generic or clichéd and may lack a distinctive voice ✓
• C. AI always produces stories that are too long
• D. AI can only write in non-English languages

Why: AI models tend to produce predictable phrasing and familiar plot patterns because they draw on common patterns in their training. Writers add value by revising for voice, specificity, and freshness.

• A. Accept the first character description the AI gives you
• B. Ask the AI for character traits, motivations, or conflicts, then build on them with your own vision ✓
• C. Let the AI decide every character's name and personality with no input
• D. Avoid character work with AI because it only creates villains

Why: Using AI to generate possibilities for traits, motivations, and conflicts can deepen characters, but the writer should shape those elements to fit the story's themes and their own understanding of the character.

• A. Publishing AI output directly without any edits
• B. Rewriting AI-suggested passages in your own words and voice ✓
• C. Asking the AI to imitate a specific living author's exact sentences
• D. Combining several AI outputs without reading them

Why: Rewriting AI suggestions in your own voice keeps the work authentically yours and helps avoid unintentionally reproducing generic or derivative text. Reading and revising everything is essential to creative ownership.

• A. Having AI replace all of your original scenes automatically
• B. Asking AI for feedback on pacing, clarity, or inconsistencies, then deciding which suggestions to apply ✓
• C. Using AI to delete any sentence you wrote that it does not understand
• D. Submitting the draft to AI and skipping your own review

Why: AI can act as a useful editing assistant by pointing out possible issues in pacing, clarity, or continuity. The writer remains in control, evaluating each suggestion against their artistic intent.

• A. Making sure the images are generated as quickly as possible
• B. Verifying that the generated images do not infringe on existing copyrighted works or the rights of artists ✓
• C. Choosing the highest resolution output available
• D. Using only the most popular AI model on the market

Why: AI image generators may be trained on copyrighted material, so creators should check that outputs do not reproduce protected works or unfairly imitate specific living artists before commercial use.

• A. Hiding the use of AI so that audiences focus only on the final product
• B. Disclosing when and how AI tools were used in creating a work ✓
• C. Making AI-generated work completely invisible to the audience
• D. Publishing the source code of every AI model used

Why: Transparency means being honest about AI's role in the creative process, which helps audiences, collaborators, and clients understand how a work was made and evaluate it accordingly.

• A. Biased data always makes AI run slower
• B. Biased training data can lead to outputs that stereotype or underrepresent certain groups of people ✓
• C. Biased data only affects text-based AI, not images
• D. Bias in data is legally required for creative tools

Why: If training data overrepresents some groups or contains stereotypes, AI-generated creative content can reproduce and amplify those stereotypes, harming representation and fairness.

• A. Prompting the AI to copy a living artist's exact signature style without credit or permission
• B. Using AI to mimic a specific artist's work and selling it as an original by that artist
• C. Developing your own artistic voice and disclosing AI assistance while avoiding direct imitation of specific living artists without consent ✓
• D. Scraping artists' portfolios to build a private training dataset without asking

Why: Respecting artists means creating original work, being open about AI use, and avoiding unauthorized imitation or exploitation of individual creators' distinctive styles and labor.

• A. AI outputs are always low quality, so they are obviously fake
• B. It can mislead audiences and spread misinformation about real events or people ✓
• C. Documentary evidence must always be typed rather than drawn
• D. There is no concern, since audiences can always tell the difference

Why: Realistic AI-generated content can be mistaken for real photos or footage, so presenting it as authentic can deceive viewers and contribute to misinformation, especially when it depicts real people or events.

• A. Asking the AI to generate the entire final product immediately
• B. Defining the project goal, audience, and constraints clearly before prompting ✓
• C. Choosing the most advanced AI tool available regardless of the task
• D. Copying a finished example from another creator without modification

Why: Clear goals, audience, and constraints give the AI the context it needs to produce relevant output and help you evaluate whether results meet your intent.

• A. Accepting the first output the AI produces without any changes
• B. Generating a draft, reviewing it, refining the prompt, and repeating until the result fits your vision ✓
• C. Using AI only once at the very end of the project to polish the work
• D. Asking the AI to start over with a completely new topic after each response

Why: Iteration, which involves drafting, critiquing, and refining, is central to AI-assisted creativity because first outputs are rarely the final version.

• A. AI-generated content is always plagiarized and must be removed
• B. AI can produce factual errors, generic phrasing, or content that does not match your intent, so human judgment is needed ✓
• C. Editing is only required when submitting work to a government agency
• D. AI output cannot be edited under any circumstances

Why: AI tools can make mistakes, default to generic language, or drift from the project's purpose, so creators must review, verify, and shape the output.

• A. Writing only the word 'bad' whenever the output is not perfect
• B. Specifying what works, what does not, and why, such as 'Make the tone more playful and shorten the second paragraph' ✓
• C. Refusing to provide any feedback and hoping the AI improves on its own
• D. Repeating the exact same prompt without any changes

Why: Specific, actionable feedback helps the AI understand what to change, leading to more targeted and useful revisions.

• A. Only the final product, with no notes on the process
• B. The prompts used, the tools applied, the revisions made, and your own creative decisions ✓
• C. The AI company's private internal data
• D. Nothing, since AI work does not need documentation

Why: Recording prompts, tools, revisions, and your own decisions makes the process transparent, reproducible, and clearly shows your creative contribution.