McAfee-Secured Website

Exam Code: AI-103

Exam Name: Developing AI Apps and Agents on Azure

Certification Provider: Microsoft

Corresponding Certification: Microsoft Certified: Azure AI Apps and Agents Developer Associate

100% Microsoft AI-103 Guaranteed Success With Testking Training Materials

Get Real AI-103 Exam Questions & Expert Verified Answers as Experienced on the Real Exam!

187 Questions & Answers with Testing Engine

"Developing AI Apps and Agents on Azure Exam", also known as AI-103 exam, is a Microsoft certification exam.

Pass your tests with the always up-to-date AI-103 Exam Engine. Your AI-103 training materials keep you at the head of the pack!

Satisfaction Guaranteed

Money Back Guarantee

Testking provides hassle-free money back guarantee with our products. That is because we have 100% trust in the abilities of our professional and experience product team, and our record is a proof of that.

99.6% PASS RATE
Was: $137.49
Now: $124.99

Microsoft AI-103 Practice Test 38

Question 370

An approved landscape image is too narrow for a panoramic website header. The team wants to extend the scenery beyond the photo edges rather than change the central composition. Which edit is designed for this?

  1. Apply a small interior inpainting mask over the main subject.
  2. Horizontally stretch the source image to the new header width without adding any scenery.
  3. Crop both sides of the photograph to reduce its width.
  4. Use outpainting to extend a photo beyond its original boundaries.

Explanation

Correct Answer

D:

Outpainting generates plausible surrounding imagery beyond the existing borders while retaining the central reference.

Incorrect Answers

A:

Interior patch replacement does not add pixels beyond the existing canvas edges.

B:

Rescaling existing pixels distorts the center content instead of synthesizing new imagery beyond original boundaries.

C:

Cropping removes content and cannot extend the image to the requested wider aspect.

Question 371

A brand team will swap only the background in an approved product photo. The product's logo and packaging shape must remain recognizable. What method fits?

  1. Use generic image captioning as the final edited asset.
  2. Regenerate the entire product from text without its reference image.
  3. Apply a global style transfer to the complete image including the logo.
  4. Use masked reference-guided editing and verify unmasked product areas.

Explanation

Correct Answer

D:

Source-guided edits limit variation outside the intended area and permit preservation checks.

Incorrect Answers

A:

Captions describe content rather than modifying the photo.

B:

Pure text generation may alter brand and product geometry.

C:

Unconstrained full-image changes can distort protected marks.

Question 372

A business analyst uploads an existing financial chart and asks where two lines cross. The app mistakenly routes the request to an image generator. What model capability is needed?

  1. Use image editing to remove axes so the intersection is visually clearer.
  2. Apply outpainting beyond the chart borders and treat new pixels as data.
  3. Use visual understanding over the supplied chart, with grounding in observed labels and plotted relationships.
  4. Generate a visually similar replacement chart and answer from its values rather than the uploaded chart.

Explanation

Correct Answer

C:

The task interprets existing graphical evidence rather than creating new imagery.

Incorrect Answers

A:

Removing axes can destroy information needed for interpretation.

B:

Canvas expansion does not reveal real plotted values.

D:

Reconstruction can change the plot; only the source chart supports a grounded answer.

Question 373

A film preproduction team has written scene descriptions but no footage. It needs a short concept clip demonstrating camera motion. Which workflow fits?

  1. Run video transcription on an empty audio file.
  2. Use a still-image caption as the complete animation file.
  3. Apply a video editing filter to footage that has not been produced.
  4. Select a text-to-video workflow for a storyboard concept.

Explanation

Correct Answer

D:

Video generation synthesizes motion from textual scene guidance when no source footage exists.

Incorrect Answers

A:

Speech recognition cannot create the requested moving imagery.

B:

Text description is not a time-based video asset.

C:

Editing assumes a source clip to transform.

Question 374

A director supplies a reference clip to preserve a character's look while generating a new short scene. Why should the generation request include the approved footage?

  1. Discard the reference and rely on a generic text prompt for exact appearance.
  2. Use supported reference-guided video generation with consistency constraints and final visual review.
  3. Use audio transcription of the reference as the sole video input.
  4. Generate each frame independently from the prompt without evaluating temporal identity consistency.

Explanation

Correct Answer

B:

Source footage can guide identity and visual continuity beyond a text description alone.

Incorrect Answers

A:

Text alone may not constrain the character consistently.

C:

A transcript omits much of the character's visual appearance.

D:

Independent frames may drift in objects and motion without temporal controls.

Question 375

An advertiser has an approved 15-second video and wants to alter the lighting of two seconds without remaking the clip. What capability should be considered?

  1. Use pure text-to-video generation with no original clip.
  2. Select a video-editing workflow for an existing clip.
  3. Regenerate an audio transcript and treat it as new video frames.
  4. Replace the complete video with a still image caption.

Explanation

Correct Answer

B:

Targeted video editing preserves the existing narrative while modifying selected footage.

Incorrect Answers

A:

Starting from scratch can change shot content and timing.

C:

Speech processing cannot directly edit lighting in visual frames.

D:

A caption cannot preserve the original moving clip.

Question 376

A generated animation of a moving vehicle changes its color and wheel count between frames. What should the editing pipeline validate?

  1. Render still images at regular intervals without enforcing subject continuity between adjacent frames.
  2. Compare consecutive frames for color, geometry and object continuity.
  3. Inspect only the first frame's resolution.
  4. Increase audio bitrate and assume image identities stabilize.

Explanation

Correct Answer

B:

Video quality depends on continuity of objects through time, not just each frame's local appearance.

Incorrect Answers

A:

A sequence of unrelated frames will not satisfy a coherent video-generation requirement.

C:

One frame cannot establish continuity across the sequence.

D:

Audio settings do not control frame-to-frame visual structure.

Question 377

An inpainting job replaces a torn label but also moves a nearby button, violating the repair specification. Which QA detects the problem?

  1. Discard the original reference after generation.
  2. Detect spatial layout drift after inpainting a specific region.
  3. Check only whether the output image file exists.
  4. Judge the edited image solely by its overall brightness.

Explanation

Correct Answer

B:

Region-aware comparison catches unintended modifications beyond the authorized patch.

Incorrect Answers

A:

Without source comparison unauthorized changes are hard to identify.

C:

File creation does not establish preservation of untouched pixels.

D:

Global brightness cannot detect a moved component.

Question 378

A social-media campaign requires portrait 9:16 assets for mobile displays, but the generator produces square images. Which setting should be planned?

  1. Rename the file with `_9x16` while leaving the pixels square.
  2. Use square generation and assume the platform will never crop important elements.
  3. Increase prompt temperature as the sole means of controlling geometry.
  4. Set a supported portrait aspect ratio or resolution matching the destination and verify safe framing.

Explanation

Correct Answer

D:

Output dimensions must match the placement while preserving essential content.

Incorrect Answers

A:

A filename cannot change actual image dimensions.

B:

Automatic cropping may remove logos or text.

C:

Sampling randomness is not an aspect-ratio specification.

Question 379

A print house needs high-resolution generated artwork for a poster, but current 512px exports show pixelation at full size. What should the design check?

  1. Reduce the model's max text-output tokens to improve print fidelity.
  2. Change the CSS width of the online preview and assume the print resolution increases.
  3. Increase JPEG compression until the image file is smaller.
  4. Choose supported generation/upscaling resolution based on the required print size and effective DPI.

Explanation

Correct Answer

D:

Pixel dimensions and output scale determine whether the printed asset meets sharpness requirements.

Incorrect Answers

A:

Language-token budgets do not set image raster resolution.

B:

Display scaling does not add detail to the original bitmap.

C:

Stronger compression can worsen visible artifacts and does not add pixels.

Question 380

A catalog team needs many color-background variants without altering the original product silhouette. Which variation policy is appropriate?

  1. Mask background edits and compare the original product silhouette.
  2. Generate each variant solely from text and accept any product shape.
  3. Apply unmasked global reshaping to maximize novelty.
  4. Validate only output file count regardless of visual content.

Explanation

Correct Answer

A:

Limiting transformations protects fixed product geometry while allowing intended background diversity.

Incorrect Answers

B:

Unanchored generation can change the advertised merchandise.

C:

Full-image changes conflict with the silhouette requirement.

D:

Quantity alone does not establish preserved product identity.

Developing AI Apps and Agents on Azure (AI-103) Practice Tests