how i make AI videos that print me $12k per day
Published
Should I watch this?
Skip to the real contentYour full summary is ready below. Read on while we finish your watch recommendation.
Your summary is ready. The watch recommendation is next.
Worth watchingSkim
The competitor research and image-to-video demo offer useful reference-image selection, action prompts, and fixes for missing product variants, but repetitive virality advice and earnings pitches add little, and the finished example has no reported sales results.
The main answer
The video’s practical answer is: create realistic AI product shorts by combining competitor research, strong reference images, an attention-grabbing opening, and carefully directed animation. It does not establish a reliable formula for virality or income. The demonstrated workflow is: 1. Collect clear product photos, ideally from multiple angles. 2. Research competing Instagram videos. Find an advantage: better visuals, a fresh concept, or a different product variation. Skip products where you cannot identify one. 3. Analyze a successful reference video in Claude using the creator’s custom video-analysis skills. Connect Claude to the generation platform called “Xfield” in the transcript through MCP. 4. Supply product, person, and background references, then generate a realistic starting image. The creator sources references from Pinterest, but that does not establish permission to reuse someone’s photo or likeness. 5. Write explicit directions for speech, movement, reaction, camera angles, and the product reveal. Generate roughly 10-second, 1080p clips; request additional clips when the output misses requirements. 6. Edit the clips together with readable text, appropriate placement, and a curiosity-building opening. The central creative principle: a convincing starting image and clear action directions improve realism, while the opening gives viewers a reason to keep watching.
Key takeaways
- The creator prioritizes a distinctive, pattern-interrupting hook over simply copying common gifting or store scenarios.
- Lighting, believable expressions, background choice, and text placement are presented as important quality controls—not independently demonstrated guarantees of reach.
- Competitor research should reveal a concrete reason your video would be worth watching instead of existing alternatives.
- The demonstration requires revisions: the first generated clip omits requested product variations.
- Views are not sales, revenue is not profit, and another seller’s conversion rate cannot establish your likely earnings.
- Use authorized reference assets and avoid presenting synthetic reactions as genuine customer experiences.
Summary
The creator teaches an AI-assisted workflow for making realistic short product-promotion videos, emphasizing competitor research, novel hooks, reference images, and precise prompts. He describes generating and assembling a product-reveal video, then pivots to student earnings claims and an invitation to join his mentorship. The transcript contains practical production advice, but no verified evidence that the demonstrated video generated sales or that beginners typically achieve the advertised results.
What’s being promoted?
Primary offer: The creator explicitly promotes his mentorship through the first link in the description. He says participants receive a “winning product,” an individualized strategy, and a system for posting frequently. Pricing, terms, support details, and typical outcomes are not disclosed. Lead generation and audience retention: Viewers must message “prompt” through his Instagram story to obtain the custom skills. This moves interested viewers into direct contact and could support later sales conversations, although the transcript does not show those conversations. Recommendations to watch his batching tutorial and a student case study also keep viewers within his content. Tools promoted or recommended: “Xfield” receives the strongest platform endorsement and repeated demonstration. Claude, SnapInsta, Pinterest, and CapCut/TikTok fonts support the workflow. A preferred video model is called the best, but its name is unclear in the transcription and no comparison substantiates that claim. Sponsorships, affiliate commissions, or ownership relationships are not disclosed in the transcript; they cannot be assumed. Persuasive framing: Large earnings figures, beginner success stories, “knowledge gap” language, and Q4 urgency position the mentorship as a shortcut to financial success. The opening estimate applies one student’s claimed $3,000-per-million-views rate to someone else’s 14.6 million views. That equals $43,800, but it is an extrapolation—not verified earnings from that video. The million-dollar month, 70% margin, and other income claims likewise lack assessable substantiation in the supplied text. Referenced screenshots are not available here, and costs, failure rates, and typical results are missing. Bottom line: This is both a tutorial and an explicit mentorship sales pitch. The commercial agenda is clear; whether the earnings claims are accurate, representative, or attributable to the coaching remains uncertain. Those gaps do not, by themselves, prove fraud.
Source: Reader-provided transcript
