The output gets the brand right in the outro card, but the mid-script phone b-roll is generic message-bubble UI rather than an actual Duolingo lesson screen, so product specificity is only partial.
What was measured
Product understanding
Whether the tool gets the product right in features and positioning, rather than being generic or inaccurate.
decisive for this rankingtransformation
An ad has to describe the product accurately and position it correctly; generic or wrong messaging breaks the output. (3 of 3 judges)
What was given, what came back
Test input: Duolingo – Mobile App · mixed · group: ugc-style-video-ads
Input — what we sent
Input, verbatim
Product URL: https://www.duolingo.com/ Exact script: Hook: Learning a new language has never been this easy. Script: I've been using Duolingo for a few minutes every day, and it's made language learning simple and fun. The short lessons are easy to follow, and the daily practice keeps me motivated. If you're planning to learn a new language, give Duolingo a try. Output requirements: - Vertical (9:16) - 20–30 seconds - UGC-style AI presenter - Natural AI voice - Captions enabled
A short-form UGC promo prompt for a consumer mobile app, designed to test whether the tool can explain app benefits simply and persuasively while generating a realistic AI-presenter video.
Why this input is hard
- · Mobile app understanding
- · Consumer app promotion
- · Script quality
- · Avatar realism
- · Voice quality
- · Caption quality
- · Overall ad readiness
Output — unretouched
Also checked on this input — same tool, 5 other criteria
Ad-readiness◐ MixedThe clip is branded and complete, but the generic b-roll and non-clickable end card make it less action-ready as a paid ad without another review pass.Avatar realism✓ WorkedThe presenter reads as a consistent human avatar with a clean UGC look throughout the clip, with no obvious flicker or identity break reported.B-roll / product integration◐ MixedThe clip adds a timed phone-and-coffee cutaway, but the screen content is blurred and generic instead of product-specific, so the supporting footage does not fully reinforce the app.Caption quality◐ MixedThe captions visibly advance across the clip and support the CTA line, but the report does not separately validate timing or styling beyond the sampled frame progression.Script quality & persuasiveness✓ WorkedThe script is reproduced verbatim and reaches the closing CTA line without truncation, so the testimonial structure and call-to-action remain intact.
Provenance
- Observation
- 7a6225f6-e4ba-40b5-b2f7-b5ee9a7ca9ba
- Evidence run
- 9fdbb84b-4aa0-484d-8f89-5c46608019a5
- Study
- Generate UGC-Style Video Ads With AI Avatars
- Research task
- 86baxegpg
- Tested at
- not recorded
- Source
- first-party
- Evidence state
- verified
- Proof shown
- input + output shown
- Cost / latency
- not captured
- Repeat run
- not captured
- Tester
- not captured
The last three rows are honest blanks, not placeholders — our capture has no field for them yet.
Query this
get_evidence({
tool: "invideo-ai",
scenario: "ugc-style-video-ads"
})MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 5 other tools
measured on Product understanding
AKOOL⚠ StruggledKeeps the language-learning message simple, but the output stays generic and does not make Duolingo feel distinctive.Creatify.ai◐ MixedFor the app input, the ad correctly sold language-learning convenience and daily practice, but the report says the background stayed generic and the Duolingo interface was not clearly shown.HeyGen✗ FailedProduct understanding breaks in the B-roll cutaway: the phone screen shows a shopping app ('NEW YORK BAG $219, ADD TO CART') instead of Duolingo or any language-learning interface.JoggAI✗ FailedFor an app prompt, the tool can leave out the app completely and produce only an avatar talking to camera, with no screenshot, graphic, or in-video product reference.Vidnoz AI✓ WorkedUnderstands a language-learning app well enough to position it around short lessons, daily practice, and simple, fun learning.
This evidence is published in
Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com