Gains:
- Ability to classify on a task basis where the artificial intelligence output is a hypothesis and where it is a reliable assistant
- Multi-layer verification of an AI-supported design in terms of production, safety, ergonomics and authenticity
- Ability to acquire the habit of using artificial intelligence in a responsible, transparent and privacy-respecting manner in the ever-changing vehicle ecosystem
Module Exam
1. A designer presents an attractive concept rendering produced by an artificial intelligence visual tool to the customer as the 'final product to be produced' without any modifications. What is the fundamental mistake in this approach?
- A) AI rendering does not guarantee manufacturability; Material, wall thickness and assembly cannot be presented as final product without verification by DFM analysis and engineer approval ✔
- B) Render's resolution is not high enough for client presentation
- C) Only the background color was not chosen according to the brand
- D) It was necessary to run the same prompt three times and present the average image
Description: Although the AI rendering is visually convincing, it carries no guarantees in terms of wall thickness, assembly, tolerance, material behavior and manufacturability; It often contains details that cannot be physically reproduced. The output is an unconfirmed hypothesis; The production decision must go through DFM analysis and engineer approval.
2. What does 'risk-based classification' enable when using artificial intelligence in industrial design?
- A) To reduce the subscription fee of the artificial intelligence tool
- B) Determine the role of artificial intelligence and the mandatory verification depth according to the risk level of the task ✔
- C) Prompts should be written shorter
- D) Automatically transfer all design decisions to artificial intelligence
Explanation: Not every design task is at the same level of risk. Classifying the task as low/medium/high/critical determines which output can be freely used and which requires material-production verification and competent expert approval.
3. Which of the following is the strongest contribution of artificial intelligence in concept production?
- A) Giving the exact dimensions of the final product to be produced
- B) Producing a wide and diverse pool of ideas/sketches in a short time and expanding the field of exploration ✔
- C) Determining exactly which concept is patentable
- D) Calculation of material strength
Explanation: The main value of artificial intelligence at the concept stage is that it expands the designer's field of exploration by providing a large number and variety of ideas (divergent production) in a short time. The choice of idea, function and manufacturability judgment belongs to the designer.
4. What is the use of fixing the 'seed' value in an image prompt?
- A) Automatically transfers the copyright of the image to the designer
- B) Providing a reproducible, consistent starting point with the same prompt ✔
- C) Doubles the resolution of the output
- D) It guarantees manufacturability
Explanation: In diffusion models, seed is the randomness point where production begins. The same prompt and the same seed usually produce a similar/same composition; This provides consistency for reproducing an image or testing small changes in a controlled manner.
5. In ergonomic decisions, artificial intelligence is asked 'What is the average hand width in mm?' Why is it risky to ask and write a single number directly to the handle size?
- A) Ergonomics works with percentile distribution and population; single 'average' number is insufficient and value should be confirmed by reliable anthropometric data, mockup/test ✔
- B) It should have been asked to artificial intelligence in inches
- C) The number is correct, there is only a rounding error
- D) Hand width is never used in the design
Explanation: Ergonomics works with a percentile distribution of the target user population (e.g. 5th–95th percentile), not a single 'average', and varies by region/population. AI can hallucinate a general number; size should be verified from reliable anthropometric data table and mockup/user testing.
6. What is the most correct attitude when artificial intelligence extracts a theme from an interview text in user research synthesis?
- A) Putting the emerging themes directly into the report
- B) Verify each theme and quote by linking it back to the raw interview data, checking for fabricated quotes ✔
- C) Only take the theme that artificial intelligence says is the strongest
- D) Deleting the raw data and keeping only the artificial intelligence summary
Explanation: AI can extract themes quickly but may create bias by making up non-existent quotes or exaggerating/deleting minority opinion. Each theme and quote should be validated by linking back to raw interview data.
7. Why is the 'draft angle' necessary in a plastic part to be produced by injection molding and what should be done when artificial intelligence suggests it?
- A) It is necessary for the part to come out of the mold without any damage; The angle suggested by artificial intelligence must be verified by material, texture and molder's opinion ✔
- B) Only necessary to determine the color of the part
- C) The angle given by artificial intelligence is universally correct for every material
- D) Drawing angle is used only on metal parts
Explanation: Drawing angle is the slight inclination given to the walls so that the part can come out of the mold without difficulty; Otherwise, the part will get stuck in the mold and the surface will be damaged. AI may suggest a typical angle range, but the actual value is verified by material, surface texture, and molder's opinion.
8. What is the most critical control before importing the output of an AI tool that generates a 3D model from text/image into the CAD workflow?
- A) Checking that the file name of the model is correct
- B) Validate the model in CAD in terms of size, scale, mesh quality and manufacturability, and remodel if necessary ✔
- C) Just look at the color and add it to the workflow
- D) Sending the model directly to production without checking it at all
Description: Although artificial intelligence-generated 3D models look good visually, they are often out of scale, have a thick/dirty mesh structure, have holes or have unproducible geometry. Actual size, scale, wall thickness and manufacturability need to be checked and remodeled in CAD.
9. Which is mandatory in terms of honesty when presenting a photorealistic product rendering produced with artificial intelligence to the customer?
- A) Clearly stating that the render is a concept visualization and the product to be produced may not be exactly the same ✔
- B) Presenting the rendering as a real photo
- C) Hiding size and material differences from the customer
- D) Never say that the visual is produced by artificial intelligence
Description: Artificial intelligence rendering may show different actual product size, material texture and detail. It is a requirement of honest representation to clearly state that the visual is a concept visualization and that the product to be produced may not be exactly the same as the rendering.
10. What is the best way to ensure consistency when producing product family images with artificial intelligence for a brand?
- A) Maintaining defined brand elements such as form signature, ratio and CMF in each variation with reference and fixed instructions and controlling the output with the brand guide ✔
- B) Producing completely random, unrelated images for each product
- C) Ignoring the brand guide at all
- D) Copying the style of a different artist each time
Description: Brand/product language consistency; It is achieved by preserving defined elements such as form signature, ratio, CMF (colour-material-surface) in each variation. Reference image, fixed style instruction and comparison with the brand guide make this possible; The designer checks each output against the guide.
11. Why is it risky to have artificial intelligence produce a product image 'in the style of a famous designer/brand' and use it in a commercial project?
- A) Registered design may cause legal risks and ethical problems in terms of copyright and unfair competition; It is necessary to develop an original direction and seek legal advice when necessary ✔
- B) Only the resolution of the image remains low
- C) There is no risk, artificial intelligence output is always unique
- D) Only rendering time increases
Explanation: Imitating an existing trademark, a registered design or the signature style of a well-known artist/designer may create legal risks in terms of design registration, copyright and unfair competition; It is also problematic in terms of originality and professional ethics. It is necessary to develop original direction and seek legal advice when necessary.
12. What is the best course of action in terms of confidentiality when handing over a design brief, customer name and confidential product details to a cloud-based AI tool?
- A) Pasting the entire brief as is with real name and confidential details
- B) Anonymize context and clear customer name and confidential product details, use corporate tool that complies with privacy/NDA ✔
- C) Assuming real customer name is essential for result quality
- D) Ignoring the confidentiality agreement at all
Disclosure: Brief, customer identity, undisclosed product concept and price/strategy information are trade secret and contract (NDA) sensitive. It is necessary to anonymize the context, clear the real name and confidential details, choose the tool that guarantees corporate/privacy, and comply with company/client policy.
13. What is the main benefit of writing 'strong prompt' instead of 'weak prompt' in visual artificial intelligence?
- A) Brings output closer to design intent, reduces cliché and increases reproducibility ✔
- B) Automatically copyright protects the output of artificial intelligence
- C) Guarantees manufacturability
- D) Always cuts rendering time in half
Description: A vague, short prompt pushes the AI into random and cliche output. A structured prompt that clarifies subject, material, angle, light, style and negative elements brings the output closer to the design intent and increases reproducibility.
14. What should be the role of AI output in a safety-critical product (e.g. a children's toy or occupational safety equipment) in industrial design?
- A) Producing only concepts and drafts; Safety compliance, strength and risk decisions are made by competent experts and real tests ✔
- B) Delegating security approval completely to artificial intelligence
- C) Skipping standard tests and relying on artificial intelligence visuals
- D) Using artificial intelligence rendering as a certification document
Description: There are risks such as injury, suffocation, and lack of strength in safety-critical products. Artificial intelligence can produce concepts and blueprints, but compliance with safety standards, strength and risk analysis are done by competent engineers/experts and real tests; AI output does not replace this approval.