{"observation":{"id":"2399684a-6762-4a39-a524-676aeddfa25c","tool":"heygen","tool_name":"HeyGen","criterion":"avatar-realism","criterion_name":"Avatar realism","criterion_definition":"Believable presenter; accurate lip-sync; how uncanny or AI-obvious the presenter looks.","criterion_evidence_type":"transformation","criterion_rank_role":"decisive","criterion_rank_role_reason":"If the presenter looks fake or lip-sync is off, the ad fails at the core UGC avatar job. (3 of 3 judges)","scenario":"futuresmart-ai-saas","scenario_name":"FutureSmart AI – SaaS","group_tag":"ugc-style-video-ads","scenario_description":"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.","modality":"mixed","input_text":"Product URL: https://aidemos.com/\n\nExact script:\nHook: Finding the right AI tools shouldn't take hours.\nScript: 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.\n\nOutput requirements:\n- Vertical (9:16)\n- 20–30 seconds\n- UGC-style AI presenter\n- Natural AI voice\n- Captions enabled","input_artifact_refs":[],"stresses":["SaaS product understanding","Script quality","Avatar realism","Voice quality","Caption quality","Overall ad readiness"],"verdict":"mixed","score":null,"score_total":null,"note":"The presenter stays visually believable overall, but the output has a fused/warped right hand at 1s and 14.9s, which breaks the realism enough to keep it from being cleanly convincing.","evidence_state":"verified","source":null,"artifacts":[{"url":"https://cdn.futuresmart.ai/public/aidemos/565abebfad4d45568f0393e660861f55.png?v=1","role":"context","alt":null},{"url":"https://cdn.futuresmart.ai/public/aidemos/04a030ae7bbc48a28ae9395f52551b8d.mp4?v=1","role":"output","alt":null}],"run_id":"9fdbb84b-4aa0-484d-8f89-5c46608019a5","study_title":"Generate UGC-Style Video Ads With AI Avatars","study_kind":"generation","research_task":"86baxegpg","tested_at":null,"completeness":"input-and-output","input":{"state":"text","text":"Product URL: https://aidemos.com/\n\nExact script:\nHook: Finding the right AI tools shouldn't take hours.\nScript: 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.\n\nOutput requirements:\n- Vertical (9:16)\n- 20–30 seconds\n- UGC-style AI presenter\n- Natural AI voice\n- Captions enabled","files":[],"modality":"mixed","stresses":["SaaS product understanding","Script quality","Avatar realism","Voice quality","Caption quality","Overall ad readiness"]},"tool_page_slug":"heygen","tool_url":"https://aidemos.com/tools/heygen","permalink":"https://aidemos.com/evidence/2399684a-6762-4a39-a524-676aeddfa25c","api_url":"https://ai.aidemos.com/v1/observations/2399684a-6762-4a39-a524-676aeddfa25c"},"peers":[{"id":"de783a96-9ea4-4004-a7d5-22d3c084cd07","tool":"akool","tool_name":"AKOOL","verdict":"mixed","score":null,"score_total":null,"note":"Produces a usable talking-head presenter, but the avatar still reads as somewhat AI-generated and less natural in parts, so realism is only moderate.","artifact_count":1,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/35e80b86028344bc874c0a681c1f3f03.mp4?v=1","evidence_url":"https://aidemos.com/evidence/de783a96-9ea4-4004-a7d5-22d3c084cd07"},{"id":"b034f0c5-4dec-47f4-8163-f99facc3b6d5","tool":"dreamface","tool_name":"DreamFace","verdict":"mixed","score":null,"score_total":null,"note":"The FutureSmart avatar is visually believable, with natural skin and hair texture, but a fused-finger hand glitch at about 0:02 breaks the realism.","artifact_count":2,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/9e9d39795e0c438b8e898c4e2db099b8.mp4?v=1","evidence_url":"https://aidemos.com/evidence/b034f0c5-4dec-47f4-8163-f99facc3b6d5"},{"id":"cd120e81-2ff0-4549-bcd0-8401cb620246","tool":"invideo-ai","tool_name":"Invideo AI","verdict":"worked","score":null,"score_total":null,"note":"The presenter reads as a believable human talking head, with natural blinking and clean hand rendering reported across the render and no flicker or warping noted.","artifact_count":2,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/093d24dd0ea0430eb36a89eeea1d9e3a.mp4?v=1","evidence_url":"https://aidemos.com/evidence/cd120e81-2ff0-4549-bcd0-8401cb620246"},{"id":"3c44a1fd-97ee-4a28-a9d1-45ad62796d1b","tool":"joggai","tool_name":"JoggAI","verdict":"worked","score":null,"score_total":null,"note":"Generates a convincing photorealistic talking-head avatar, with the report noting natural skin, hair, teeth, expressive gesturing, and no visible deformities or robotic feel.","artifact_count":2,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/4b156d6342724ed1b57be3d863fab95e.mp4?v=1","evidence_url":"https://aidemos.com/evidence/3c44a1fd-97ee-4a28-a9d1-45ad62796d1b"},{"id":"276e8d59-7732-4c6c-be12-a6c451b9f166","tool":"synthesia","tool_name":"Synthesia","verdict":"worked","score":null,"score_total":null,"note":"The FutureSmart output used a stable talking-head presenter that stayed centered through the short video, and the report says lip-sync was consistent across the run.","artifact_count":1,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/6217146be0434680a0d9e8add56290da.mp4?v=1","evidence_url":"https://aidemos.com/evidence/276e8d59-7732-4c6c-be12-a6c451b9f166"},{"id":"43756801-68fd-4a57-a791-d21835bd556f","tool":"topview-ai","tool_name":"Topview AI","verdict":"mixed","score":null,"score_total":null,"note":"The tool can keep a vertical presenter visible throughout the clip, but the presenter still reads as AI-generated: the report says realism was lower than higher-end tools and expression/gesture range was minimal.","artifact_count":2,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/57a22e612d4b431391560092245658e8.mp4?v=1","evidence_url":"https://aidemos.com/evidence/43756801-68fd-4a57-a791-d21835bd556f"},{"id":"3658a77d-4eab-4199-859b-dfdbacef2fa5","tool":"vidnoz-ai","tool_name":"Vidnoz AI","verdict":"mixed","score":null,"score_total":null,"note":"Produces 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.","artifact_count":1,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/a634e30078db49f3b9014626950d0fa1.mp4?v=1","evidence_url":"https://aidemos.com/evidence/3658a77d-4eab-4199-859b-dfdbacef2fa5"}],"other_criteria":[{"id":"fc836d16-388e-4ee4-bec0-cb53fc29b6f0","criterion":"caption-quality","criterion_name":"Caption quality","rank_role":"decisive","verdict":"failed","score":null,"score_total":null,"note":"Captioning 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.","artifact_count":1,"evidence_url":"https://aidemos.com/evidence/fc836d16-388e-4ee4-bec0-cb53fc29b6f0"},{"id":"d85a401c-453e-41da-8da1-d013b88303b0","criterion":"platform-formats","criterion_name":"Platform formats","rank_role":"context","verdict":"failed","score":null,"score_total":null,"note":"The output only fills about 55% of the 9:16 canvas, leaving large dead white bars top and bottom, so the clip is not a native full-bleed vertical export.","artifact_count":3,"evidence_url":"https://aidemos.com/evidence/d85a401c-453e-41da-8da1-d013b88303b0"}],"appears_in":[{"page_type":"ranking","slug":"ugc-video-ad-generators","title":"Best AI Tools for Generating UGC-Style Video Ads With AI Avatars","url":"https://aidemos.com/best/ugc-video-ad-generators","binding":"run"}],"same_scenario":[{"id":"de783a96-9ea4-4004-a7d5-22d3c084cd07","tool":"akool","tool_name":"AKOOL","verdict":"mixed","score":null,"score_total":null,"note":"Produces a usable talking-head presenter, but the avatar still reads as somewhat AI-generated and less natural in parts, so realism is only moderate."},{"id":"b034f0c5-4dec-47f4-8163-f99facc3b6d5","tool":"dreamface","tool_name":"DreamFace","verdict":"mixed","score":null,"score_total":null,"note":"The FutureSmart avatar is visually believable, with natural skin and hair texture, but a fused-finger hand glitch at about 0:02 breaks the realism."},{"id":"cd120e81-2ff0-4549-bcd0-8401cb620246","tool":"invideo-ai","tool_name":"Invideo AI","verdict":"worked","score":null,"score_total":null,"note":"The presenter reads as a believable human talking head, with natural blinking and clean hand rendering reported across the render and no flicker or warping noted."},{"id":"3c44a1fd-97ee-4a28-a9d1-45ad62796d1b","tool":"joggai","tool_name":"JoggAI","verdict":"worked","score":null,"score_total":null,"note":"Generates a convincing photorealistic talking-head avatar, with the report noting natural skin, hair, teeth, expressive gesturing, and no visible deformities or robotic feel."},{"id":"276e8d59-7732-4c6c-be12-a6c451b9f166","tool":"synthesia","tool_name":"Synthesia","verdict":"worked","score":null,"score_total":null,"note":"The FutureSmart output used a stable talking-head presenter that stayed centered through the short video, and the report says lip-sync was consistent across the run."},{"id":"43756801-68fd-4a57-a791-d21835bd556f","tool":"topview-ai","tool_name":"Topview AI","verdict":"mixed","score":null,"score_total":null,"note":"The tool can keep a vertical presenter visible throughout the clip, but the presenter still reads as AI-generated: the report says realism was lower than higher-end tools and expression/gesture range was minimal."},{"id":"3658a77d-4eab-4199-859b-dfdbacef2fa5","tool":"vidnoz-ai","tool_name":"Vidnoz AI","verdict":"mixed","score":null,"score_total":null,"note":"Produces 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."}]}