Can generate burned-in subtitles and lets the user edit them after creation, but the caption text can be awkward—the preview shows a duplicated opening phrase, 'Welcome to FutureSmart AI AI to discover'.
What was measured
Caption quality
Accurate, well-timed captions with appropriate, on-trend styling.
decisive for this rankingtransformation
Accurate, well-timed captions are part of the finished ad and affect clarity and performance on social platforms. (3 of 3 judges)
What was given, what came back
Test input: FutureSmart AI – SaaS · mixed · group: ugc-style-video-ads
Input — what we sent
Input, verbatim
Product URL: https://aidemos.com/ Exact script: Hook: Finding the right AI tools shouldn't take hours. Script: I've been using FutureSmart AI to discover and compare AI tools in one place. It helps me find the right tool faster with real use cases, rankings, and detailed comparisons. If you regularly use AI tools for work, it's definitely worth checking out. Output requirements: - Vertical (9:16) - 20–30 seconds - UGC-style AI presenter - Natural AI voice - Captions enabled
A short-form UGC ad prompt for a SaaS product discovery/comparison platform, designed to test whether the tool understands a software product, writes a believable testimonial-style script, and produces a realistic AI presenter video with captions.
Why this input is hard
- · SaaS product understanding
- · Script quality
- · Avatar realism
- · Voice quality
- · Caption quality
- · Overall ad readiness
Output — unretouched




Also checked on this input — same tool, 6 other criteria
Ad-readiness◐ MixedProduces a usable base ad, but the slightly AI-obvious presenter and structured, less-natural delivery mean it does not look fully polished enough to run unchanged as a paid ad.Avatar realism◐ MixedProduces a presentable talking-head presenter, but the avatar still reads as slightly AI-generated rather than fully human; the report calls the realism only moderate.Editing/refinement✓ WorkedSupports post-generation refinement without full regeneration: the ad can be reopened, the script/subtitles edited in place, and the updated text is reflected back in the editor.Product understanding✓ WorkedUnderstands a product-discovery SaaS well enough to frame it around comparing AI tools in one place, real use cases, rankings, and detailed comparisons instead of generic software praise.Script quality & persuasiveness◐ MixedWrites a clear testimonial-style promo, but the delivery is still template-like and more structured than natural UGC, so the hook and CTA feel only moderately persuasive as an ad.Voice quality✓ WorkedGenerates a clear, stable spoken delivery that the report says is understandable and suitable for informational content.
Provenance
- Observation
- 76b72ebd-55b4-429f-8a68-9216fd07beca
- 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: "vidnoz-ai",
scenario: "ugc-style-video-ads"
})MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 5 other tools
measured on Caption quality
AKOOL✗ FailedDoes not automatically include captions in the initial output; the preview showed no captions during the tested workflow.HeyGen✗ FailedCaptioning is absent in the FutureSmart run: the researcher checked 8 timestamps from 0s through 14.9s and found zero burned-in text anywhere in the clip.Invideo AI✓ WorkedThe captions advance progressively and stay readable, and the report says they are well-timed to speech up to the cutoff.JoggAI✗ FailedThe tool can export a clip with zero on-screen captions for the full 18.4 seconds, so captioning is not baked into the output.Topview AI◐ MixedCaptions were generated and generally aligned with speech, but the run had minor timing issues.
This evidence is published in
Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com