{"content_id":"itbeipwc6o","slug":"ai-youtube-side-hustle-workflow-case-study","locale":"en","schema_type":"Article","category":"case_study","category_name":"Case Study","title":"Case Study: How a Worker in Their 50s Produces and Monetizes AI YouTube Videos as a Side Hustle","summary":"This analysis examines an interview with a worker in their 50s who used AI to produce and monetize long-form science videos. It explains the reality of the 1–2-hour production process and outlines key risks that must be reviewed, including earnings verification, copyright, repetitive or reused content, and synthetic content disclosure.","author":{"name":"Injoys Editorial Team","url":"https://injoys.com/ko/about"},"key_points":["The interviewee said that an AI-based production process reduced the time required to create each video to 1–2 hours, but this is an individual case shaped by personal proficiency and tool configuration, not a universal benchmark.","The reported cumulative earnings of approximately KRW 17 million and monthly net profit of KRW 4.5–5 million are self-reported by the interviewee and have not been independently verified through YouTube Analytics, payment statements, expense records, or tax documents.","Using AI does not itself prohibit monetization, but mass-produced repetitive content or reused content that lacks substantive commentary may affect monetization across the entire channel.","More important than production speed are the originality of the topic, source verification, the commentary and perspective provided to viewers, thumbnail accuracy, and final human review.","YouTube's disclosure requirements for altered or synthetic content may apply to realistic-looking synthetic scenes or content that meaningfully alters real people or events."],"content_markdown":"AI can be used to quickly create scripts, narration, images, and rough edits. However, “producing a video in 2 hours” and “earning 5 million won per month” are separate issues. The former concerns production efficiency, while the latter is a business outcome determined by views, advertising rates, viewer countries, video length, costs, taxes, and other factors.\n\nThis article reconstructs the production process using interview material from an IT employee in his 50s as a case study. The earnings figures are self-reported by the interviewee and are neither externally verified averages nor reproducible income guarantees.\n\n## Case Overview\n\nThe interviewee said that while considering his income and lifestyle after retirement, he came across a free YouTube-related course and started a side business. He initially considered economics and science as potential topics, then focused on the long-form science channel that achieved monetization results first.\n\n| Item | Information provided in the interview | Points to consider when interpreting |\n|---|---|---|\n| Primary job | IT-related role | Familiarity with technology may have reduced the initial setup time |\n| Content | Long-form videos on science topics | Fact-checking and the accuracy of visual materials are especially important |\n| Production frequency | Uploading almost every day | Differentiation and quality for each video take priority over frequency |\n| Production time | Approximately 1–2 hours per video | Time spent on topic research, error correction, and tool setup may have been excluded |\n| Cumulative earnings | 17 million won over approximately 3 months after uploads began | Self-reported by the interviewee; no analytics screens or payment records were provided |\n| Monthly earnings | Explained as approximately $3,000 per month, or approximately 4.5–5 million won after taxes | Exchange rates, platform withholding, domestic taxes, and tool costs must be distinguished |\n| Highest daily earnings | Approximately 610,000 won | A peak does not represent average daily earnings or a sustainable level |\n\nThe material states that video uploads began in April, but the year is not specified. Therefore, the results should not be assumed to belong to a particular year.\n\n## How to Interpret the “5 Million Won per Month” Figure\n\n### Revenue, Payouts, and Net Profit Are Different\n\nEstimated revenue displayed in YouTube Studio may not be the same as the final payout or net business profit. To compare actual performance, the following items must be separated.\n\n- **Estimated advertising revenue:** The amount provisionally displayed in YouTube Analytics\n- **Finalized earnings:** The amount confirmed as payable after adjustments are applied\n- **Platform payout:** The amount actually deposited after withholding or payment adjustments\n- **Net business profit:** The payout minus AI subscription fees, voice and video generation API costs, music and source material costs, equipment expenses, and taxes\n\nBased on the material alone, it is impossible to determine whether the “4.5–5 million won per month after taxes” mentioned in the interview refers to the amount after platform withholding or net profit after domestic comprehensive income tax. The statements about cumulative earnings of 17 million won and approximately $3,000 each month are also difficult to reconcile precisely without view counts and exchange-rate data for each period.\n\n### Data Needed to Verify Performance\n\nAt a minimum, the following metrics are needed to evaluate an earnings case.\n\n1. The exact period during which the earnings were generated\n2. Monthly views and valid playbacks\n3. RPM and playback-based CPM\n4. Viewers’ primary countries and languages\n5. The share of revenue by source, including long-form advertising revenue, YouTube Premium, and affiliate revenue\n6. Estimated revenue before cancellations and adjustments, and the actual payout\n7. Costs for AI tools, APIs, music, and outsourced work\n8. A distinction between pre-tax and after-tax profit\n\nA channel experiencing a rapid short-term surge in views may record high peak daily earnings, but there is no guarantee that the same level will continue afterward.\n\n## Reconstructing the AI Video Production Process\n\nThe process described by the interviewee is closer to a semi-automated workflow that connects multiple generation tools to automate rough production while leaving final decisions to a person, rather than an approach in which AI operates the entire channel without human involvement.\n\n### Step 1: Finding Potential Topics\n\nThe interviewee explained that he subscribes to multiple science channels and looks for videos whose recent views are increasing rapidly relative to their subscriber counts. This can be used as a signal of viewers’ current interests, but it does not mean that the structure and expression of the original video may be copied.\n\nSafer research methods include the following.\n\n- Find questions or phenomena that have recently gained attention.\n- Review papers, public agency materials, and multiple explanatory sources rather than relying on a single video.\n- Design a core question, an original explanatory sequence, and a visual approach that differ from the original.\n- Confirm that the title and thumbnail match the actual content of the video.\n- Record the source and verification date for each referenced claim in the working document.\n\nA “video with relatively few subscribers but many views” is only a clue for discovering topic demand. Imitating the title, script progression, and thumbnail composition of a particular video may create copyright and reused-content issues.\n\n### Step 2: Compiling Source Materials and Writing the Script\n\nThe interviewee said that he provides a reference video URL to Claude and requests a script approximately 10 minutes long to create the basic structure. However, it should not be assumed that AI can always accurately read the full video, subtitles, or latest information from a URL. The scope of processing may vary depending on access permissions, subtitle availability, and service functionality.\n\nA more reliable method is to organize and provide verified materials first.\n\n- The core question to explain and the target audience\n- Facts confirmed through authoritative sources\n- Speculation and unverified figures that must not be used\n- The video’s original perspective or analogy\n- The desired length, writing style, and scene structure\n- Instructions that each fact must be rechecked\n\nOnce the AI draft is complete, a person must recheck proper nouns, dates, figures, causal relationships, and the level of scientific consensus. Any quotations or research findings not found in the sources must be removed.\n\n### Step 3: Generating Voice and Visual Materials\n\nIn this case, voice, image, and video generation services were connected to an integrated production solution. The process involved creating narration with a voice API such as ElevenLabs and generating photorealistic images and short intro videos.\n\nThe following risks must be reviewed at this stage.\n\n- Whether a real person’s voice or face has been imitated without permission\n- Whether generated images distort scientific facts\n- Whether actual observational images could be confused with conceptual illustrations\n- Whether the rights to logos, characters, photos, or video sources are being infringed\n- Whether realistic-looking synthetic scenes require disclosure\n- Whether the conditions for commercial use of music and sound effects are met\n\nIn science videos, generated images may serve as explanatory visualizations. If they are presented as though they were actual microscope images, astronomical observation data, or historical records, it is safer to clearly explain within the video that they are synthetic images.\n\n### Step 4: Reviewing the Rough Edit in CapCut\n\nOnce the integrated tool creates a project file with the script, narration, images, and video placed on the timeline, subtitles, background music, scene lengths, and other elements are adjusted in CapCut. Even after automatic placement is complete, a person must review the following from beginning to end.\n\n- Whether the narration and visuals explain the same facts\n- Whether numbers, units, and proper nouns in subtitles are accurate\n- Whether scene transitions are excessively repetitive\n- Whether voice pronunciation and sentence pacing sound natural\n- Whether the background music conflicts with the narration\n- Whether materials requiring attribution are presented in a misleading way\n- Whether each video offers substantive commentary and new value\n\n### Step 5: Final Human Approval Before Uploading\n\nThe interviewee said that even when an automatic upload feature is available, he reviews the video himself before publishing it. This is an important control procedure for reducing errors, rights violations, and incorrect publication settings.\n\nBefore uploading, confirm the following.\n\n1. Whether the title and thumbnail exaggerate or misrepresent the video’s content\n2. Whether the necessary source and rights information has been included in the description\n3. Whether the made-for-kids setting has been configured correctly\n4. Whether realistic altered or synthetic content must be disclosed\n5. Whether there are any issues with copyright check results or music licenses\n6. Whether someone is responsible for reviewing comments and viewer reports after publication\n\n## Conditions That Make 1–2 Hour Production Possible\n\nProducing one video in 1–2 hours is not a standard time that every creator can achieve immediately. It is more appropriate to view this as a compressed editing time made possible when the following prerequisites are in place.\n\n- A reusable project template is available.\n- Voice, image, and editing APIs have already been integrated.\n- The channel’s writing style and visual style have been established.\n- The creator understands the tools’ error patterns and how to correct them.\n- Time for topic research and collecting reference materials is handled separately.\n- Complex animation or original filming is not required.\n\nAccurate science content may require more time for research and verification than for production itself. Omitting fact-checking to save time can undermine trust and monetization potential in the long run.\n\n## Points to Consider Under YouTube Monetization Policies\n\n### The Nature of the Result Matters More Than AI Use Itself\n\nYouTube does not exclude every video from monetization solely because it was created with AI. However, content that uses templates to mass-produce nearly identical videos or lacks substantive differences between videos may be considered repetitive or mass-produced content.\n\nReviewers may examine not only individual videos but also the channel’s main themes, most-viewed videos, newest videos, share of watch time, and metadata such as titles, thumbnails, and descriptions. Therefore, simply adding a logo or changing the subtitle font does not guarantee originality.\n\n### Reused Content and Copyright Are Separate Issues\n\nEven if no copyright claim is filed or permission has been obtained when using edited footage from another creator, the video may still fail to meet YouTube’s reused-content criteria. Conversely, claiming to have added original commentary does not automatically grant the right to use the material.\n\nA video containing reused materials should be able to clearly answer the following questions.\n\n- What new explanation or analysis does it provide compared with the original?\n- Is it evident that the creator personally planned and narrated the video?\n- Do viewers have an independent reason to watch this video instead of the original?\n- Are the necessary rights secured for the video, images, and music used?\n\n### Disclosure of Synthetic Content\n\nIf realistic-looking synthetic content alters the words or actions of a real person, modifies a real place or event, or depicts a realistic scene that never occurred, it may need to be disclosed as altered or synthetic content during the upload process. Simple script assistance, subtitle generation, and clearly unrealistic visual effects may not be treated at the same level, so the official criteria and specific context must be reviewed.\n\nMaking a disclosure does not automatically restrict monetization. However, repeatedly failing to make required disclosures may cause YouTube to apply labels or impose penalties.\n\n## Reviewing the Claim That “Improving Quality Prevents Channel Deletion”\n\nThe interview material includes advice suggesting that adding personal opinions and a logo to improve quality nearly eliminates the risk of penalties. This should not be accepted as a guaranteed rule.\n\nA channel’s status is determined by considering the following factors together.\n\n- Whether the Community Guidelines have been violated\n- Whether copyright infringement has occurred and warnings have accumulated\n- Whether spam, deception, or manipulation has occurred\n- Whether the content is repetitive or reused under monetization policies\n- Whether synthetic-content disclosure requirements have been followed\n- The overall context of the videos, titles, thumbnails, and descriptions\n\nSuperficial changes such as adding a logo, background music, or one or two sentences of personal opinion are not sufficient. The key factors are original research, independent structure, meaningful commentary, accurate information, and human editorial judgment.\n\n## Sustainable Operating Metrics\n\nLooking only at upload volume and revenue may obscure a channel’s vulnerabilities. It is advisable to track the following metrics together.\n\n| Area | Metrics to review |\n|---|---|\n| Viewer response | Impressions click-through rate, average view duration, audience retention, returning viewers |\n| Content quality | Number of factual corrections, missing sources, viewer correction requests, number of original materials per video |\n| Revenue | RPM, monthly finalized earnings, tool costs, profit per video, revenue source concentration |\n| Production | Time spent on research, scripting, generation, and review stages; rework rate; generation failure costs |\n| Policies and rights | Copyright claims, warnings, limited ads, synthetic-content disclosure status |\n\nRather than repeating the superficial topics of profitable videos, creators should analyze the explanatory approaches that produced high retention and the questions viewers asked.\n\n## Actionable Safety-Focused Workflow\n\n### Weekly Planning\n\n- Gather 10 potential topics and check search demand and recency.\n- Confirm that each potential topic has sufficient reliable sources.\n- Choose a question or explanatory approach that differs from existing popular videos.\n\n### Production for Each Video\n\n- Read the materials and create a list of facts and a source record.\n- Use AI to generate an outline and draft.\n- Have a person verify claims, figures, and wording, and add perspective.\n- Add voice, images, video, and music for which usage rights have been confirmed.\n- Review the automatically edited result from beginning to end.\n- Add any required synthetic-content disclosures and information in the description.\n\n### Post-Publication Management\n\n- Do not make thumbnails provocative based solely on the initial click-through rate.\n- Review errors identified by viewers and make corrections if necessary.\n- Record both the cost and finalized earnings for each video.\n- Review the channel once per month to determine whether it is becoming excessively repetitive.\n\n## Realistic Conclusions from the Case\n\nThe central point of this case is not that AI automatically generates 5 million won per month. Rather, automation reduced the time spent on script drafts, voice generation, image creation, and rough editing, allowing the creator to devote the remaining time to topic selection, thumbnails, and final review.\n\nHowever, the interview earnings were not independently verified, and using the same tools and upload frequency does not guarantee the same results. The factors that increase the likelihood of success are not simply enduring for 3 months, but conducting accurate research, providing original commentary, complying with policies, managing costs, and continuously improving based on viewer data.","content_html":"\u003cp\u003eAI can be used to quickly create scripts, narration, images, and rough edits. However, “producing a video in 2 hours” and “earning 5 million won per month” are separate issues. The former concerns production efficiency, while the latter is a business outcome determined by views, advertising rates, viewer countries, video length, costs, taxes, and other factors.\u003c/p\u003e\n\u003cp\u003eThis article reconstructs the production process using interview material from an IT employee in his 50s as a case study. The earnings figures are self-reported by the interviewee and are neither externally verified averages nor reproducible income guarantees.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#case-overview\" class=\"anchor\" id=\"case-overview\"\u003e\u003c/a\u003eCase Overview\u003c/h2\u003e\n\u003cp\u003eThe interviewee said that while considering his income and lifestyle after retirement, he came across a free YouTube-related course and started a side business. He initially considered economics and science as potential topics, then focused on the long-form science channel that achieved monetization results first.\u003c/p\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eItem\u003c/th\u003e\n\u003cth\u003eInformation provided in the interview\u003c/th\u003e\n\u003cth\u003ePoints to consider when interpreting\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003ePrimary job\u003c/td\u003e\n\u003ctd data-label=\"Information provided in the interview\"\u003eIT-related role\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eFamiliarity with technology may have reduced the initial setup time\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eContent\u003c/td\u003e\n\u003ctd data-label=\"Information provided in the interview\"\u003eLong-form videos on science topics\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eFact-checking and the accuracy of visual materials are especially important\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eProduction frequency\u003c/td\u003e\n\u003ctd data-label=\"Information provided in the interview\"\u003eUploading almost every day\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eDifferentiation and quality for each video take priority over frequency\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eProduction time\u003c/td\u003e\n\u003ctd data-label=\"Information provided in the interview\"\u003eApproximately 1–2 hours per video\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eTime spent on topic research, error correction, and tool setup may have been excluded\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eCumulative earnings\u003c/td\u003e\n\u003ctd data-label=\"Information provided in the interview\"\u003e17 million won over approximately 3 months after uploads began\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eSelf-reported by the interviewee; no analytics screens or payment records were provided\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eMonthly earnings\u003c/td\u003e\n\u003ctd data-label=\"Information provided in the interview\"\u003eExplained as approximately $3,000 per month, or approximately 4.5–5 million won after taxes\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eExchange rates, platform withholding, domestic taxes, and tool costs must be distinguished\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Item\"\u003eHighest daily earnings\u003c/td\u003e\n\u003ctd data-label=\"Information provided in the interview\"\u003eApproximately 610,000 won\u003c/td\u003e\n\u003ctd data-label=\"Points to consider when interpreting\"\u003eA peak does not represent average daily earnings or a sustainable level\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003cp\u003eThe material states that video uploads began in April, but the year is not specified. Therefore, the results should not be assumed to belong to a particular year.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#how-to-interpret-the-5-million-won-per-month-figure\" class=\"anchor\" id=\"how-to-interpret-the-5-million-won-per-month-figure\"\u003e\u003c/a\u003eHow to Interpret the “5 Million Won per Month” Figure\u003c/h2\u003e\n\u003ch3\u003e\n\u003ca href=\"#revenue-payouts-and-net-profit-are-different\" class=\"anchor\" id=\"revenue-payouts-and-net-profit-are-different\"\u003e\u003c/a\u003eRevenue, Payouts, and Net Profit Are Different\u003c/h3\u003e\n\u003cp\u003eEstimated revenue displayed in YouTube Studio may not be the same as the final payout or net business profit. To compare actual performance, the following items must be separated.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eEstimated advertising revenue:\u003c/strong\u003e The amount provisionally displayed in YouTube Analytics\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eFinalized earnings:\u003c/strong\u003e The amount confirmed as payable after adjustments are applied\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003ePlatform payout:\u003c/strong\u003e The amount actually deposited after withholding or payment adjustments\u003c/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eNet business profit:\u003c/strong\u003e The payout minus AI subscription fees, voice and video generation API costs, music and source material costs, equipment expenses, and taxes\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eBased on the material alone, it is impossible to determine whether the “4.5–5 million won per month after taxes” mentioned in the interview refers to the amount after platform withholding or net profit after domestic comprehensive income tax. The statements about cumulative earnings of 17 million won and approximately $3,000 each month are also difficult to reconcile precisely without view counts and exchange-rate data for each period.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#data-needed-to-verify-performance\" class=\"anchor\" id=\"data-needed-to-verify-performance\"\u003e\u003c/a\u003eData Needed to Verify Performance\u003c/h3\u003e\n\u003cp\u003eAt a minimum, the following metrics are needed to evaluate an earnings case.\u003c/p\u003e\n\u003col\u003e\n\u003cli\u003eThe exact period during which the earnings were generated\u003c/li\u003e\n\u003cli\u003eMonthly views and valid playbacks\u003c/li\u003e\n\u003cli\u003eRPM and playback-based CPM\u003c/li\u003e\n\u003cli\u003eViewers’ primary countries and languages\u003c/li\u003e\n\u003cli\u003eThe share of revenue by source, including long-form advertising revenue, YouTube Premium, and affiliate revenue\u003c/li\u003e\n\u003cli\u003eEstimated revenue before cancellations and adjustments, and the actual payout\u003c/li\u003e\n\u003cli\u003eCosts for AI tools, APIs, music, and outsourced work\u003c/li\u003e\n\u003cli\u003eA distinction between pre-tax and after-tax profit\u003c/li\u003e\n\u003c/ol\u003e\n\u003cp\u003eA channel experiencing a rapid short-term surge in views may record high peak daily earnings, but there is no guarantee that the same level will continue afterward.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#reconstructing-the-ai-video-production-process\" class=\"anchor\" id=\"reconstructing-the-ai-video-production-process\"\u003e\u003c/a\u003eReconstructing the AI Video Production Process\u003c/h2\u003e\n\u003cp\u003eThe process described by the interviewee is closer to a semi-automated workflow that connects multiple generation tools to automate rough production while leaving final decisions to a person, rather than an approach in which AI operates the entire channel without human involvement.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#step-1-finding-potential-topics\" class=\"anchor\" id=\"step-1-finding-potential-topics\"\u003e\u003c/a\u003eStep 1: Finding Potential Topics\u003c/h3\u003e\n\u003cp\u003eThe interviewee explained that he subscribes to multiple science channels and looks for videos whose recent views are increasing rapidly relative to their subscriber counts. This can be used as a signal of viewers’ current interests, but it does not mean that the structure and expression of the original video may be copied.\u003c/p\u003e\n\u003cp\u003eSafer research methods include the following.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eFind questions or phenomena that have recently gained attention.\u003c/li\u003e\n\u003cli\u003eReview papers, public agency materials, and multiple explanatory sources rather than relying on a single video.\u003c/li\u003e\n\u003cli\u003eDesign a core question, an original explanatory sequence, and a visual approach that differ from the original.\u003c/li\u003e\n\u003cli\u003eConfirm that the title and thumbnail match the actual content of the video.\u003c/li\u003e\n\u003cli\u003eRecord the source and verification date for each referenced claim in the working document.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eA “video with relatively few subscribers but many views” is only a clue for discovering topic demand. Imitating the title, script progression, and thumbnail composition of a particular video may create copyright and reused-content issues.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#step-2-compiling-source-materials-and-writing-the-script\" class=\"anchor\" id=\"step-2-compiling-source-materials-and-writing-the-script\"\u003e\u003c/a\u003eStep 2: Compiling Source Materials and Writing the Script\u003c/h3\u003e\n\u003cp\u003eThe interviewee said that he provides a reference video URL to Claude and requests a script approximately 10 minutes long to create the basic structure. However, it should not be assumed that AI can always accurately read the full video, subtitles, or latest information from a URL. The scope of processing may vary depending on access permissions, subtitle availability, and service functionality.\u003c/p\u003e\n\u003cp\u003eA more reliable method is to organize and provide verified materials first.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eThe core question to explain and the target audience\u003c/li\u003e\n\u003cli\u003eFacts confirmed through authoritative sources\u003c/li\u003e\n\u003cli\u003eSpeculation and unverified figures that must not be used\u003c/li\u003e\n\u003cli\u003eThe video’s original perspective or analogy\u003c/li\u003e\n\u003cli\u003eThe desired length, writing style, and scene structure\u003c/li\u003e\n\u003cli\u003eInstructions that each fact must be rechecked\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eOnce the AI draft is complete, a person must recheck proper nouns, dates, figures, causal relationships, and the level of scientific consensus. Any quotations or research findings not found in the sources must be removed.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#step-3-generating-voice-and-visual-materials\" class=\"anchor\" id=\"step-3-generating-voice-and-visual-materials\"\u003e\u003c/a\u003eStep 3: Generating Voice and Visual Materials\u003c/h3\u003e\n\u003cp\u003eIn this case, voice, image, and video generation services were connected to an integrated production solution. The process involved creating narration with a voice API such as ElevenLabs and generating photorealistic images and short intro videos.\u003c/p\u003e\n\u003cp\u003eThe following risks must be reviewed at this stage.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eWhether a real person’s voice or face has been imitated without permission\u003c/li\u003e\n\u003cli\u003eWhether generated images distort scientific facts\u003c/li\u003e\n\u003cli\u003eWhether actual observational images could be confused with conceptual illustrations\u003c/li\u003e\n\u003cli\u003eWhether the rights to logos, characters, photos, or video sources are being infringed\u003c/li\u003e\n\u003cli\u003eWhether realistic-looking synthetic scenes require disclosure\u003c/li\u003e\n\u003cli\u003eWhether the conditions for commercial use of music and sound effects are met\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eIn science videos, generated images may serve as explanatory visualizations. If they are presented as though they were actual microscope images, astronomical observation data, or historical records, it is safer to clearly explain within the video that they are synthetic images.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#step-4-reviewing-the-rough-edit-in-capcut\" class=\"anchor\" id=\"step-4-reviewing-the-rough-edit-in-capcut\"\u003e\u003c/a\u003eStep 4: Reviewing the Rough Edit in CapCut\u003c/h3\u003e\n\u003cp\u003eOnce the integrated tool creates a project file with the script, narration, images, and video placed on the timeline, subtitles, background music, scene lengths, and other elements are adjusted in CapCut. Even after automatic placement is complete, a person must review the following from beginning to end.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eWhether the narration and visuals explain the same facts\u003c/li\u003e\n\u003cli\u003eWhether numbers, units, and proper nouns in subtitles are accurate\u003c/li\u003e\n\u003cli\u003eWhether scene transitions are excessively repetitive\u003c/li\u003e\n\u003cli\u003eWhether voice pronunciation and sentence pacing sound natural\u003c/li\u003e\n\u003cli\u003eWhether the background music conflicts with the narration\u003c/li\u003e\n\u003cli\u003eWhether materials requiring attribution are presented in a misleading way\u003c/li\u003e\n\u003cli\u003eWhether each video offers substantive commentary and new value\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3\u003e\n\u003ca href=\"#step-5-final-human-approval-before-uploading\" class=\"anchor\" id=\"step-5-final-human-approval-before-uploading\"\u003e\u003c/a\u003eStep 5: Final Human Approval Before Uploading\u003c/h3\u003e\n\u003cp\u003eThe interviewee said that even when an automatic upload feature is available, he reviews the video himself before publishing it. This is an important control procedure for reducing errors, rights violations, and incorrect publication settings.\u003c/p\u003e\n\u003cp\u003eBefore uploading, confirm the following.\u003c/p\u003e\n\u003col\u003e\n\u003cli\u003eWhether the title and thumbnail exaggerate or misrepresent the video’s content\u003c/li\u003e\n\u003cli\u003eWhether the necessary source and rights information has been included in the description\u003c/li\u003e\n\u003cli\u003eWhether the made-for-kids setting has been configured correctly\u003c/li\u003e\n\u003cli\u003eWhether realistic altered or synthetic content must be disclosed\u003c/li\u003e\n\u003cli\u003eWhether there are any issues with copyright check results or music licenses\u003c/li\u003e\n\u003cli\u003eWhether someone is responsible for reviewing comments and viewer reports after publication\u003c/li\u003e\n\u003c/ol\u003e\n\u003ch2\u003e\n\u003ca href=\"#conditions-that-make-12-hour-production-possible\" class=\"anchor\" id=\"conditions-that-make-12-hour-production-possible\"\u003e\u003c/a\u003eConditions That Make 1–2 Hour Production Possible\u003c/h2\u003e\n\u003cp\u003eProducing one video in 1–2 hours is not a standard time that every creator can achieve immediately. It is more appropriate to view this as a compressed editing time made possible when the following prerequisites are in place.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eA reusable project template is available.\u003c/li\u003e\n\u003cli\u003eVoice, image, and editing APIs have already been integrated.\u003c/li\u003e\n\u003cli\u003eThe channel’s writing style and visual style have been established.\u003c/li\u003e\n\u003cli\u003eThe creator understands the tools’ error patterns and how to correct them.\u003c/li\u003e\n\u003cli\u003eTime for topic research and collecting reference materials is handled separately.\u003c/li\u003e\n\u003cli\u003eComplex animation or original filming is not required.\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eAccurate science content may require more time for research and verification than for production itself. Omitting fact-checking to save time can undermine trust and monetization potential in the long run.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#points-to-consider-under-youtube-monetization-policies\" class=\"anchor\" id=\"points-to-consider-under-youtube-monetization-policies\"\u003e\u003c/a\u003ePoints to Consider Under YouTube Monetization Policies\u003c/h2\u003e\n\u003ch3\u003e\n\u003ca href=\"#the-nature-of-the-result-matters-more-than-ai-use-itself\" class=\"anchor\" id=\"the-nature-of-the-result-matters-more-than-ai-use-itself\"\u003e\u003c/a\u003eThe Nature of the Result Matters More Than AI Use Itself\u003c/h3\u003e\n\u003cp\u003eYouTube does not exclude every video from monetization solely because it was created with AI. However, content that uses templates to mass-produce nearly identical videos or lacks substantive differences between videos may be considered repetitive or mass-produced content.\u003c/p\u003e\n\u003cp\u003eReviewers may examine not only individual videos but also the channel’s main themes, most-viewed videos, newest videos, share of watch time, and metadata such as titles, thumbnails, and descriptions. Therefore, simply adding a logo or changing the subtitle font does not guarantee originality.\u003c/p\u003e\n\u003ch3\u003e\n\u003ca href=\"#reused-content-and-copyright-are-separate-issues\" class=\"anchor\" id=\"reused-content-and-copyright-are-separate-issues\"\u003e\u003c/a\u003eReused Content and Copyright Are Separate Issues\u003c/h3\u003e\n\u003cp\u003eEven if no copyright claim is filed or permission has been obtained when using edited footage from another creator, the video may still fail to meet YouTube’s reused-content criteria. Conversely, claiming to have added original commentary does not automatically grant the right to use the material.\u003c/p\u003e\n\u003cp\u003eA video containing reused materials should be able to clearly answer the following questions.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eWhat new explanation or analysis does it provide compared with the original?\u003c/li\u003e\n\u003cli\u003eIs it evident that the creator personally planned and narrated the video?\u003c/li\u003e\n\u003cli\u003eDo viewers have an independent reason to watch this video instead of the original?\u003c/li\u003e\n\u003cli\u003eAre the necessary rights secured for the video, images, and music used?\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3\u003e\n\u003ca href=\"#disclosure-of-synthetic-content\" class=\"anchor\" id=\"disclosure-of-synthetic-content\"\u003e\u003c/a\u003eDisclosure of Synthetic Content\u003c/h3\u003e\n\u003cp\u003eIf realistic-looking synthetic content alters the words or actions of a real person, modifies a real place or event, or depicts a realistic scene that never occurred, it may need to be disclosed as altered or synthetic content during the upload process. Simple script assistance, subtitle generation, and clearly unrealistic visual effects may not be treated at the same level, so the official criteria and specific context must be reviewed.\u003c/p\u003e\n\u003cp\u003eMaking a disclosure does not automatically restrict monetization. However, repeatedly failing to make required disclosures may cause YouTube to apply labels or impose penalties.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#reviewing-the-claim-that-improving-quality-prevents-channel-deletion\" class=\"anchor\" id=\"reviewing-the-claim-that-improving-quality-prevents-channel-deletion\"\u003e\u003c/a\u003eReviewing the Claim That “Improving Quality Prevents Channel Deletion”\u003c/h2\u003e\n\u003cp\u003eThe interview material includes advice suggesting that adding personal opinions and a logo to improve quality nearly eliminates the risk of penalties. This should not be accepted as a guaranteed rule.\u003c/p\u003e\n\u003cp\u003eA channel’s status is determined by considering the following factors together.\u003c/p\u003e\n\u003cul\u003e\n\u003cli\u003eWhether the Community Guidelines have been violated\u003c/li\u003e\n\u003cli\u003eWhether copyright infringement has occurred and warnings have accumulated\u003c/li\u003e\n\u003cli\u003eWhether spam, deception, or manipulation has occurred\u003c/li\u003e\n\u003cli\u003eWhether the content is repetitive or reused under monetization policies\u003c/li\u003e\n\u003cli\u003eWhether synthetic-content disclosure requirements have been followed\u003c/li\u003e\n\u003cli\u003eThe overall context of the videos, titles, thumbnails, and descriptions\u003c/li\u003e\n\u003c/ul\u003e\n\u003cp\u003eSuperficial changes such as adding a logo, background music, or one or two sentences of personal opinion are not sufficient. The key factors are original research, independent structure, meaningful commentary, accurate information, and human editorial judgment.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#sustainable-operating-metrics\" class=\"anchor\" id=\"sustainable-operating-metrics\"\u003e\u003c/a\u003eSustainable Operating Metrics\u003c/h2\u003e\n\u003cp\u003eLooking only at upload volume and revenue may obscure a channel’s vulnerabilities. It is advisable to track the following metrics together.\u003c/p\u003e\n\u003cdiv class=\"overflow-x-auto\"\u003e\u003ctable\u003e\n\u003cthead\u003e\n\u003ctr\u003e\n\u003cth\u003eArea\u003c/th\u003e\n\u003cth\u003eMetrics to review\u003c/th\u003e\n\u003c/tr\u003e\n\u003c/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Area\"\u003eViewer response\u003c/td\u003e\n\u003ctd data-label=\"Metrics to review\"\u003eImpressions click-through rate, average view duration, audience retention, returning viewers\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Area\"\u003eContent quality\u003c/td\u003e\n\u003ctd data-label=\"Metrics to review\"\u003eNumber of factual corrections, missing sources, viewer correction requests, number of original materials per video\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Area\"\u003eRevenue\u003c/td\u003e\n\u003ctd data-label=\"Metrics to review\"\u003eRPM, monthly finalized earnings, tool costs, profit per video, revenue source concentration\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Area\"\u003eProduction\u003c/td\u003e\n\u003ctd data-label=\"Metrics to review\"\u003eTime spent on research, scripting, generation, and review stages; rework rate; generation failure costs\u003c/td\u003e\n\u003c/tr\u003e\n\u003ctr\u003e\n\u003ctd data-label=\"Area\"\u003ePolicies and rights\u003c/td\u003e\n\u003ctd data-label=\"Metrics to review\"\u003eCopyright claims, warnings, limited ads, synthetic-content disclosure status\u003c/td\u003e\n\u003c/tr\u003e\n\u003c/tbody\u003e\n\u003c/table\u003e\u003c/div\u003e\n\u003cp\u003eRather than repeating the superficial topics of profitable videos, creators should analyze the explanatory approaches that produced high retention and the questions viewers asked.\u003c/p\u003e\n\u003ch2\u003e\n\u003ca href=\"#actionable-safety-focused-workflow\" class=\"anchor\" id=\"actionable-safety-focused-workflow\"\u003e\u003c/a\u003eActionable Safety-Focused Workflow\u003c/h2\u003e\n\u003ch3\u003e\n\u003ca href=\"#weekly-planning\" class=\"anchor\" id=\"weekly-planning\"\u003e\u003c/a\u003eWeekly Planning\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003eGather 10 potential topics and check search demand and recency.\u003c/li\u003e\n\u003cli\u003eConfirm that each potential topic has sufficient reliable sources.\u003c/li\u003e\n\u003cli\u003eChoose a question or explanatory approach that differs from existing popular videos.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3\u003e\n\u003ca href=\"#production-for-each-video\" class=\"anchor\" id=\"production-for-each-video\"\u003e\u003c/a\u003eProduction for Each Video\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003eRead the materials and create a list of facts and a source record.\u003c/li\u003e\n\u003cli\u003eUse AI to generate an outline and draft.\u003c/li\u003e\n\u003cli\u003eHave a person verify claims, figures, and wording, and add perspective.\u003c/li\u003e\n\u003cli\u003eAdd voice, images, video, and music for which usage rights have been confirmed.\u003c/li\u003e\n\u003cli\u003eReview the automatically edited result from beginning to end.\u003c/li\u003e\n\u003cli\u003eAdd any required synthetic-content disclosures and information in the description.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch3\u003e\n\u003ca href=\"#post-publication-management\" class=\"anchor\" id=\"post-publication-management\"\u003e\u003c/a\u003ePost-Publication Management\u003c/h3\u003e\n\u003cul\u003e\n\u003cli\u003eDo not make thumbnails provocative based solely on the initial click-through rate.\u003c/li\u003e\n\u003cli\u003eReview errors identified by viewers and make corrections if necessary.\u003c/li\u003e\n\u003cli\u003eRecord both the cost and finalized earnings for each video.\u003c/li\u003e\n\u003cli\u003eReview the channel once per month to determine whether it is becoming excessively repetitive.\u003c/li\u003e\n\u003c/ul\u003e\n\u003ch2\u003e\n\u003ca href=\"#realistic-conclusions-from-the-case\" class=\"anchor\" id=\"realistic-conclusions-from-the-case\"\u003e\u003c/a\u003eRealistic Conclusions from the Case\u003c/h2\u003e\n\u003cp\u003eThe central point of this case is not that AI automatically generates 5 million won per month. Rather, automation reduced the time spent on script drafts, voice generation, image creation, and rough editing, allowing the creator to devote the remaining time to topic selection, thumbnails, and final review.\u003c/p\u003e\n\u003cp\u003eHowever, the interview earnings were not independently verified, and using the same tools and upload frequency does not guarantee the same results. The factors that increase the likelihood of success are not simply enduring for 3 months, but conducting accurate research, providing original commentary, complying with policies, managing costs, and continuously improving based on viewer data.\u003c/p\u003e\n","tags":["AI YouTube","YouTube side hustle","Video automation","Content monetization","Employee side hustles"],"faqs":[{"question":"Can YouTube videos made with AI be monetized?","answer":"They may be eligible. The use of AI itself is not a blanket reason for denying monetization, but repetitive content mass-produced using templates or reused content lacking substantive commentary may be excluded from monetization. Each video should demonstrate original research, structure, commentary, and editing value."},{"question":"Can a single video really be made in 1–2 hours?","answer":"Creators with templates, API integrations, an established style, and proficiency with the tools can complete the initial production and editing within 1–2 hours. However, the actual total time spent may be longer when topic research, fact-checking, initial setup, and troubleshooting are included."},{"question":"Can I earn 5 million won per month in just three months, as in this case?","answer":"There is no guarantee. The earnings in this case were self-reported by the interviewee and have not been independently verified. Actual earnings vary greatly depending on views, viewer countries, advertising demand, RPM, video length, policy status, and production costs."},{"question":"Is it safe to put the URL of a popular video into AI and have it rewrite the script?","answer":"It cannot be said to be safe. Following the original's wording, structure, or key scenes too closely may cause copyright or reused content issues. Popular videos should be used only as references for gauging demand for a topic, and multiple primary sources should be researched to develop an independent set of questions and explanatory structure."},{"question":"Does obtaining the original creator's permission eliminate reused content issues?","answer":"No. Copyright permission and eligibility for YouTube monetization are separate matters. Even with permission to use the content, it may be considered reused content if meaningful commentary, analysis, or transformative value has not been added to the original."},{"question":"Do I always have to disclose that I used AI-generated images or AI voices?","answer":"The same disclosure requirements do not apply to every type of production assistance. Disclosure may be required for meaningful synthetic content that appears realistic, such as content that alters the words or actions of a real person or modifies real events or places. You should check the specific context of the video and YouTube's latest standards."},{"question":"Can I avoid monetization penalties by adding a channel logo and personal opinions?","answer":"Simply adding a logo or brief opinions does not guarantee compliance with the policies. The channel as a whole should demonstrate original research and structure, substantive commentary, accurate information, usage rights, and human editorial judgment."},{"question":"Can YouTube's estimated revenue be treated as net profit?","answer":"No. Estimated revenue may be adjusted, and to calculate actual business net profit, AI subscription fees, API usage fees, music and source material costs, equipment costs, and taxes must be deducted. The amount paid after platform withholding must also be distinguished from domestic after-tax net profit."},{"question":"What are the most important review items for a science AI channel?","answer":"Figures, dates, units, causal relationships, and the level of scientific consensus should be verified against primary sources. You should also check that generated images are not mistaken for actual observational data and that the script does not fabricate nonexistent studies or quotations."}],"sources":[{"url":"https://support.google.com/youtube/answer/1311392?hl=ko","title":"YouTube Channel Monetization Policies","type":"source"},{"url":"https://support.google.com/youtube/answer/14328491?hl=ko","title":"YouTube Help on Disclosing Altered or Synthetic Content","type":"source"},{"url":"https://support.google.com/youtube/answer/72851?hl=ko","title":"YouTube Partner Program Overview and Eligibility Requirements","type":"source"},{"url":"https://support.google.com/youtube/answer/2797466?hl=ko","title":"YouTube Copyright Information","type":"source"}],"images":[{"id":348,"url":"https://injoys.com/rails/active_storage/blobs/proxy/eyJfcmFpbHMiOnsiZGF0YSI6NDA2OSwicHVyIjoiYmxvYl9pZCJ9fQ==--3b5091747d56e0833bafbc6c20dd1bafa8493555/ai-95500ff1.webp","is_representative":true,"generation_method":"ai_image","license":"ai_generated","mime_type":"image/webp","translations":{"ko":{"alt":"책상에 앉아 AI 기반 영상 제작·검수 과정을 살펴보는 중년 직장인 일러스트","caption":"중년 직장인이 AI를 활용해 영상 콘텐츠를 제작하고 검수하는 과정을 보여준다.","description":null},"en":{"alt":"Middle-aged office worker viewing an AI-assisted video creation and review workflow at his desk","caption":"A middle-aged worker uses AI tools to create, edit, and review video content.","description":null},"ja":{"alt":"机でAIを活用した動画の制作・確認工程を見る中年会社員のイラスト","caption":"中年の会社員がAIで動画コンテンツを制作し、確認する流れを表している。","description":null},"es":{"alt":"Trabajador de mediana edad ante un flujo de creación y revisión de videos con IA","caption":"Un trabajador de mediana edad usa IA para crear, editar y revisar contenido de video.","description":null},"id":{"alt":"Pekerja paruh baya mengamati alur pembuatan dan pemeriksaan video berbantuan AI","caption":"Seorang pekerja paruh baya memakai AI untuk membuat, mengedit, dan memeriksa konten video.","description":null},"pt":{"alt":"Profissional de meia-idade diante de um fluxo de criação e revisão de vídeos com IA","caption":"Um profissional de meia-idade usa IA para criar, editar e revisar conteúdo em vídeo.","description":null},"zh-hant":{"alt":"中年上班族坐在書桌前查看AI輔助的影片製作與審核流程","caption":"一名中年上班族運用AI工具製作、剪輯並審核影片內容。","description":null},"de":{"alt":"Mittelalter Büroangestellter betrachtet einen KI-gestützten Ablauf zur Videoerstellung und Prüfung","caption":"Ein mittelalter Büroangestellter nutzt KI, um Videoinhalte zu erstellen, zu bearbeiten und zu prüfen.","description":null}}},{"id":349,"url":"https://injoys.com/rails/active_storage/blobs/proxy/eyJfcmFpbHMiOnsiZGF0YSI6NDA3NSwicHVyIjoiYmxvYl9pZCJ9fQ==--f0e7b4da49292bfbf379aa7ad7ca2a16d9b68c96/ai-5e978cbe.webp","is_representative":false,"generation_method":"ai_image","license":"ai_generated","mime_type":"image/webp","translations":{"ko":{"alt":"영상 썸네일과 수익·저작권·AI 요소의 균형을 돋보기로 살피는 중년 직장인 일러스트","caption":"50대 직장인이 AI 유튜브 부업의 수익성과 저작권 문제를 검토하고 있다.","description":null},"en":{"alt":"Middle-aged worker examining videos and a scale balancing revenue, copyright, and AI","caption":"A worker in his 50s evaluates the revenue potential and copyright risks of an AI YouTube side business.","description":null},"ja":{"alt":"動画一覧と収益・著作権・AIのバランスを虫眼鏡で調べる中年会社員のイラスト","caption":"50代の会社員がAIを活用したYouTube副業の収益性と著作権リスクを検討している。","description":null},"es":{"alt":"Trabajador de mediana edad analiza videos y una balanza con ingresos, derechos de autor e IA","caption":"Un trabajador de unos 50 años evalúa la rentabilidad y los riesgos de autor de un canal de YouTube con IA.","description":null},"id":{"alt":"Pekerja paruh baya meneliti video dan neraca berisi pendapatan, hak cipta, serta AI","caption":"Pekerja berusia 50-an menilai potensi pendapatan dan risiko hak cipta dari usaha sampingan YouTube berbasis AI.","description":null},"pt":{"alt":"Profissional de meia-idade analisa vídeos e uma balança com receita, direitos autorais e IA","caption":"Um profissional na faixa dos 50 anos avalia o potencial de receita e os riscos autorais de um canal no YouTube com IA.","description":null},"zh-hant":{"alt":"中年上班族用放大鏡檢視影片，以及收益、版權與AI元素之間的平衡","caption":"一名50多歲的上班族正在評估AI YouTube副業的獲利潛力與版權風險。","description":null},"de":{"alt":"Mann mittleren Alters prüft Videos und eine Waage mit Einnahmen, Urheberrecht und KI","caption":"Ein Berufstätiger in den Fünfzigern bewertet Ertragspotenzial und Urheberrechtsrisiken eines KI-YouTube-Nebenjobs.","description":null}}}],"published_at":"2026-07-29T23:19:12+09:00","updated_at":"2026-07-29T23:19:12+09:00","license":"cc_by","translation_status":"reviewed","available_locales":["ko","en","ja","es"],"data_locales":["ko","en","ja","es","id","pt","zh-hant","de"],"url":"https://injoys.com/en/articles/ai-youtube-side-hustle-workflow-case-study"}