I Made a YouTube Shorts Channel Using AI (Realistic Results)
TLDRIn this video, the creator challenges themselves to make three YouTube Shorts channels using AI in 24 hours. After selecting niches like entertainment, sports, and video games, AI tools assist in creating content and finding clips. The creator uses AI to streamline content creation, reducing time-consuming tasks like searching for clips. They also tweak videos to improve engagement and avoid copyright issues. After uploading, the results are mixed, with some videos performing well while others falter. The creator plans to continue the experiment for 30 days to see if AI can drive consistent success.
Takeaways
- 😀 The creator used AI to build three YouTube Shorts channels within 24 hours.
- 🔎 Choosing the right niche is crucial to avoid low views or copyright strikes.
- 📊 The creator focused on three niches: drama and celebrity news, basketball, and video game streamers.
- 🤖 AI tools helped the creator find content, including an AI tool called 'Clip Anything' to automate finding and clipping videos.
- 🛠️ Tools like ChatGPT were used to generate channel names, while MidJourney and Canva were used for logos and banners.
- 🎬 Finding the right clips from hours of content is tedious, so AI was essential in speeding up the process.
- 🔄 The creator used CapCut to edit videos, applying filters, adding voiceovers, and tweaking to increase retention.
- ⚠️ The creator warns about copyright risks and emphasizes the importance of editing to avoid them.
- 📉 Early results showed mixed performance, with one channel receiving low engagement but potential for recovery.
- 📈 After continued uploads, one video received 14k views, showing the strategy can work over time with consistency.
Q & A
What was the main challenge in creating YouTube Shorts channels using AI?
-The main challenge was that most AI tools were not capable enough to meet the creator's needs. However, they received access to a beta tool on August 8th that allowed them to proceed.
What are the three niches the creator chose for the YouTube Shorts channels?
-The three niches chosen were drama and celebrity for entertainment, basketball for sports, and live streamers for video games.
What was the most time-consuming part of creating content for the YouTube Shorts channels?
-The most time-consuming part was finding good clips for the videos, which involved staying updated with celebrity drama, live streamers, and NBA games.
Which AI tool did the creator use to solve the issue of finding relevant clips?
-The creator used an AI tool called 'Clip Anything,' which helped find and extract the best clips by detecting visuals, facial expressions, and on-screen text.
Why is the 'line of death' in YouTube analytics a problem for shorts?
-The 'line of death' signifies that a YouTube short has stopped gaining views and is no longer being recommended by the algorithm.
What engagement rate is required for YouTube Shorts to be recommended by the algorithm?
-An engagement rate of 80% is typically required for a YouTube Short to be recommended by the algorithm.
What tools did the creator use to create the logos and banners for the channels?
-The creator used AI tools like 'MidJourney' to generate logos and Canva/Photoshop to create banners.
How did the creator enhance the videos to avoid copyright issues?
-The creator applied filters, added voiceovers, hooks, transitions, and music to make the videos more unique and reduce the risk of copyright strikes.
What were the initial results from the three different channels?
-The first two channels had low engagement and were not picked up by the algorithm. However, the third channel continued to gain views and showed potential for more growth.
What are the creator's future plans for the experiment?
-The creator plans to continue uploading shorts for 30 days to gather more data and test the effectiveness of their approach.
Outlines
🎯 Challenge to Create Three AI-Based YouTube Channels in 24 Hours
The speaker sets a personal challenge to create three different YouTube short channels using AI tools within 24 hours. They reflect on past struggles with existing AI tools, explaining that most didn't meet their needs. However, on August 8th, they gained access to a new tool that finally made their idea feasible. The first step is defining a niche, as choosing the wrong one could lead to low views, monetization issues, or copyright strikes. After researching with tools like Perplexity and ChatGPT, they narrowed the options down using three key criteria: personal interest, viral potential, and ease of finding copyright-safe content. After careful consideration, the speaker chose three niches: drama celebrities for entertainment, basketball for sports, and live streamers for video games.
🤖 Using AI Tools to Build YouTube Channels
The speaker begins building the three YouTube channels using AI. They use ChatGPT to generate 20 channel names, MidJourney to create logos, and Photoshop for simple banners. The biggest challenge is sourcing relevant clips for each niche, as it requires hours of watching content, which they don't have time for. To solve this, they use a tool called ClipAnything, which uses AI to analyze visuals, facial expressions, and screen activity to extract the best clips based on prompts. This allows the speaker to quickly gather content for their three channels, significantly reducing manual effort.
⚙️ Tweaking Content for Higher Engagement and Copyright Safety
Once clips are gathered, the speaker moves on to refining the videos for better retention and copyright safety. They suggest downloading clips via Opus Clip, then using an editor like CapCut to apply filters, add voiceovers, hooks, overlays, and transitions to make the content more unique and less likely to be flagged for copyright. The speaker acknowledges that fully editing each clip is time-consuming, but suggests the more it's done, the lower the risk of copyright issues. They upload the finished shorts to YouTube and wait for results.
📊 Reviewing YouTube Channel Performance After Initial Uploads
Three days later, the speaker reviews the performance of their shorts. The first channel had 560 views but showed the 'line of death,' meaning the short stopped getting traction. The engagement rate was low, at only 16 seconds of watch time on a 55-second video. The second channel fared slightly better, with 605 views and a 63% engagement rate, giving it more potential. The third channel's short was still gaining views, with 500 views and a 52% engagement rate. Although initial results were mixed, the speaker continues uploading more shorts and later sees success with one video hitting 14k views and others reaching 10k views. They consider continuing the experiment for 30 days to gather more data and potentially improve results.
Mindmap
Keywords
💡YouTube Shorts
💡AI tools
💡Niche
💡Viral potential
💡Copyright
💡Clip Anything
💡Retention rate
💡Opus Clip
💡CapCut
💡Engagement rate
Highlights
Challenged to create three different YouTube Shorts channels using AI within 24 hours.
Struggled with AI tools for months, but found a breakthrough on August 8th with a beta tool.
First step: defining a niche, narrowing down from over 17 niches and hundreds of sub-niches.
Criteria for niche selection: personal interest, viral potential, and ease of finding risk-free content.
Removed niches like movies and business due to copyright concerns and difficulty in sourcing content.
Final niches chosen: celebrity drama, basketball, and live streamers.
Used AI tools like ChatGPT for generating channel names and MidJourney for creating logos.
Biggest challenge: finding and clipping content, solved by using an AI tool called Clip Anything.
Clip Anything identifies the best moments using visual and audio cues, saving time.
Editing process involves tools like Opus Clip and CapCut to overlay filters, add hooks, and avoid copyright.
Initial results: Laughing Legends channel received 560 views but faced the 'line of death' (shorts stopped getting views).
Second channel received 605 views with a higher engagement rate but also faced the line of death.
Third channel was still gaining views with a 52% engagement rate, but not high enough for recommendation.
Follow-up uploads after the video recording: videos getting between 2K to 14K views.
Plans to continue the experiment for 30 days to gather more data and insights.