Let’s face it. AI is everywhere. BUT we need to understand our robot overlords if we want them to work for us and not us for them. 🤖

Many creators fear that the rise of AI somehow cheapens creativity. What if we could use it to be MORE creative? Mike Russell joins us to share how he’s using AI to be a better creator.

Don’t let the robots win.

How AI is Transforming Content Creation: Insights from Expert Mike Russell

Today, we’re diving into the magical world of AI with the help of a true sorcerer in the realm, Mike Russell. If you’ve ever fantasized about AI doing your heavy lifting, stick around because we’re about to unpack some spellbinding secrets!

Meet Mike Russell: Resident AI Wiz

Let’s spin a tale about a guy who’s been ahead of the curve for over two decades. Meet Mike Russell, not just your average tech enthusiast but a true pioneer at the intersection of audio, video, and now, AI technology. His new YouTube channel, ‘Creator Magic,’ is more than just a tech hub—it’s a masterclass in leveraging AI as core ally for creatives. Here, Mike doesn’t just use AI; he shows us all how it can magnify our creative visions, making each project sharper, more engaging, and infinitely more captivating. Through ‘Creator Magic,’ Mike demystifies complex AI processes, turning them into practical, accessible tools that anyone can use to enhance their content.

During our chat, Mike revealed how he employs AI to not only streamline content creation but also to push the boundaries of creativity and engagement. From practical AI tools that simplify the creative process to addressing the challenges and ethical considerations of technology in media, we’ll explore the strategies that any creator can adopt to enhance their digital storytelling.

How AI Optimized Mike Russell’s YouTube Venture, ‘Creator Magic’

Launching a new channel in today’s crowded digital arena can be challenging, but Mike harnessed AI as a powerful tool right from the start. AI wasn’t just a technical aid for Mike; it was a pivotal force in the creative process, helping to generate unique content ideas and develop a distinctive visual brand for ‘Creator Magic’.

Mike used AI to fill in knowledge gaps and spark ideas for videos, ensuring that each piece of content was both innovative and relevant. For the visual elements, AI played a crucial role in crafting an engaging brand identity.

Inspired by a whimsical, wizard-themed concept, Mike aimed to blend the mystical with the technological for his channel. This vision came to life in the channel’s branding, featuring enchanting thumbnails with Mike dressed as a wizard, surrounded by magical tools. 

He created these dynamic thumbnails and a striking channel logo entirely with AI, using face-swapping technology to further personalize the visuals. This unique touch—placing his image in the visuals—boosted viewer engagement and helped ‘Creator Magic’ stand out in a competitive space.

Key AI Tools Powering Mike Russell’s Process

Mike leverages AI brilliantly in his creative process for ‘Creator Magic’, ensuring his content is both innovative and engaging. Here are some specific ways he integrates AI tools:

Content Ideation with AI:

• Grok: Mike uses Grok to stay ahead of the latest discussions and emerging trends specifically relevant to AI and technology. Grok’s ability to provide real-time updates and summarize key points makes it invaluable for identifying timely and relevant topics. It not only summarizes but also sources its information, allowing Mike to trace back to the original posts for verification. 

• Product Hunt: Mike regularly explores Product Hunt to stay updated on the newest tech gadgets and software. This platform is crucial for identifying emerging tools and applications that are resonating within the tech community.

• ChatGPT: After Mike does his research, he takes his list into ChatGPT for further refinement and structuring. He inputs the selected topics and tools into ChatGPT, which assesses their relevance and potential impact, providing insights and ratings to help prioritize which ones to feature. Mike uses this feedback to structure his content, generating bullet points that outline key features and uses of each tool.

• Perplexity: Perplexity has become Mike’s go-to tool for more targeted and precise content research. It excels in pulling in detailed, specific information about complex topics. For instance, when researching a microphone like the Neumann U87, Mike used Perplexity to gather comprehensive details from various sources. This tool not only compiles exhaustive data but also cites its sources, enhancing the reliability and trustworthiness of the information provided.

• Balancing AI and Human Creativity:

• Content Evaluation: Mike reminds us of the importance of verifying the accuracy of AI suggestions to ensure the reliability of the content, emphasizing the critical role of human oversight in the creative process.

• Guideline, Not Script: Although AI helps in generating initial content outlines, Mike uses these AI-generated bullet points as guidelines, not scripts. He leans into human creativity during his presentations, riffing off the points with his own imput to maintain a natural and engaging delivery.

• Navigating AI Limitations:

• Quality Control: Mike candidly mentions that not all AI tools meet his standards, referring to some as “garbage.” 

• Human Oversight: He reiterates the importance of not relying entirely on AI for content creation but instead using it to augment and enhance the creative process.

It’s clear that AI has significantly streamlined Mike’s research process, contributing to the success of his channel. However, it’s his genuine passion and interest in technology that truly resonate with his audience. While the initial draw may be the depth of his research, it’s Mike’s charismatic delivery and enthusiasm for technology that keep viewers coming back for more.

Analyzing Audience Engagement with Precision Tools

Mike combines standard analytics and advanced AI technologies to gain a deeper understanding of how audiences interact with his YouTube channel, ‘Creator Magic’. While he utilizes YouTube Studio for foundational insights into viewer behaviors, he leverages more sophisticated tools like Assembly AI’s Universal-1 transcription model to take his analytics a step further. This tool not only transcribes spoken content but also provides sentiment analysis, offering a detailed breakdown of audience reactions. Such insights are invaluable, enabling Mike to pinpoint the content that most resonates with his viewers and tailor his video creation more effectively.

Mike’s analytical process doesn’t stop at sentiment analysis. He meticulously reviews YouTube Studio metrics to identify peaks in viewer engagement, noting which segments hold audience attention the most. Armed with this data, Mike downloads transcripts segmented by speaker through Adobe Premiere Pro, focusing on these high-engagement areas. He then uses AI platforms such as ChatGPT or Claude to transform these key moments into engaging headlines and promotional materials, effectively enhancing his social media strategies. Additionally, he experiments with thumbnail optimization based on these insights, conducting A/B tests to determine which thumbnails increase watch time. By iterating on thumbnails based on observed viewer engagement, he ensures that his content not only attracts viewers but keeps them watching longer.

Balancing AI and Expertise in Video Editing

Mike blends his seasoned editor’s expertise with cutting-edge AI technology, ensuring his storytelling remains both authentic and captivating. He kicks off his editing wizardry in Riverside FM for initial edits, like removing unnecessary pauses and filler words. However, when it comes to more detailed editing decisions, Mike prefers manual oversight. He has experimented with various automated editing tools but often finds their suggestions lack the nuanced judgment required for high-quality output. Instead, he takes a hands-on approach, especially during the final editing phases, applying his expertise to ensure the content’s flow and emotional impact are strong. Although AI proposes edits and identifies potential content, Mike finds these recommendations sometimes miss the mark, stripping away elements that add depth and interest to the narrative. His approach shows that even in an era of advanced tech, the human touch is what truly makes content come alive, proving that some things just can’t be left to algorithms.

Harnessing AI for Creative Inspiration

In a world saturated with technology, Mike views AI as a potent tool to combat the dreaded “blank page syndrome” that every creator faces. Mike candidly shares his journey, noting the challenges of consistently generating fresh and engaging content for his new channel. To spark creativity, he often turns to AI language models like ChatGPT or Claude, asking them for content suggestions tailored to his channel’s theme. This initial step offers a springboard of ideas which he then refines and aligns with his unique voice.

While AI provides a starting point, Mike emphasizes the importance of injecting personal insight into the process. He’s found that AI can sometimes produce generic or irrelevant content recommendations because it draws from a vast but impersonal dataset. To counter this, Mike advises being as specific as possible when interacting with AI, treating it like “the best intern you’ll ever have.” By providing detailed instructions and lots of context—such as past video transcripts—creators can guide the AI to generate ideas that are not only relevant but also resonate in their authentic voice.

Mike also reflects on the broader implications of AI in creativity, contrasting the initial excitement surrounding AI tools with the reality of their current capabilities. He notes the importance of maintaining a human touch, especially in direct communications like newsletters, where personal quirks and genuine interaction are key to connecting with an audience. Ultimately, while AI can streamline the ideation process and aid in logistical tasks, Mike champions the value of personal creativity and authenticity, ensuring his content remains distinctly human and relatable.

AI for Beginners and the Pitfalls to Avoid

Mike starts by addressing the common pitfall many new creators fall into: being dazzled by “shiny tools” that promise to simplify content creation. He stresses the importance of not relying solely on AI to drive content. However, where AI does excell in content creation is by streamlining the research process. For instance, where he once might have visited several websites to gather comprehensive details about a product, AI tools can now seamlessly compile large amounts of varied data into cohesive summaries. This is particularly useful when you need exhaustive details on specific topics, allowing creators to focus more on content delivery rather than preliminary research.

Fact-Checking AI’s Output

Mike pointed out how important it is to rigorously fact-check AI-generated content to avoid the pitfalls of misinformation—commonly known as AI ‘hallucinations.’ He references the Gemini model by Google, designed to flag potential inaccuracies by color-coding them, helping creators distinguish reliable from questionable content. This level of scrutiny is crucial, especially when AI tools, despite their advances, still struggle with accuracy and context.

Navigating Commercial Safety in AI-Generated Content

He emphasizes the critical importance of commercial safety when using AI tools, especially for agencies and large corporations that must navigate strict legal boundaries. While such organizations cannot afford to risk deploying content that may be legally questionable, Mike notes that individual creators, although under less stringent constraints, should also be cautious. He advises creators to think ahead and consider the long-term implications of their choices, as AI-generated content can sometimes blur legal lines.

Staying informed on the developers behind AI tools can help creators evaluate whether these platforms are trustworthy and worth any risks involved. For example, OpenAI’s DALL-E has made a commitment to stand behind their users should legal issues arise. This level of support is particularly valuable as creators face the complex interplay of innovation and regulation in AI-driven content creation.

When it comes to visual content creation, Mike draws attention to the varying degrees of legal safety offered by different platforms. For example, while Midjourney provides incrediblele creative capabilities, its compliance with copyright laws remains uncertain, which could expose users to future legal challenges. In contrast, Adobe’s Firefly is designed with ‘commercial safety’ in mind, offering users the confidence that their content adheres to copyright norms and is supported by Adobe’s robust legal framework.

We also covered AI-driven audio creation tools, such as Udio and Suno. These platforms are revolutionizing music production by enabling users to generate complex music tracks using AI. However, they too face legal challenges, as their algorithms are often trained on copyrighted materials without explicit permissions. The ongoing legal battles and negotiations with major music labels exemplify the complexities of copyright law in the age of AI-enhanced creative tools.

In essence, while AI can significantly streamline content creation across various media, Mike advocates for a balanced approach where human creativity guides AI efficiency. This ensures not only that the content remains original and engaging but also that it adheres to evolving legal standards, safeguarding creators from potential legal repercussions in a rapidly advancing digital world.

The Future of AI

Mike shares his perspective on the evolving role of AI in content creation over the next year with a mix of anticipation and realism. He acknowledges the difficulty in predicting exact advancements due to the rapid evolution of technology but is hopeful about the increasing sophistication of AI tools. 

He foresees a significant shift towards AI integration that might eventually make manual content creation optional. This possibility points to a future where AI could handle routine or repetitive creative tasks, allowing human creators to focus on more nuanced aspects of their work or to innovate in new areas. Mike is particularly intrigued by the prospect of AI avatars that could represent creators around the clock, answering audience queries and performing tasks that currently require constant human attention. He imagines these avatars as specialized entities that are deeply knowledgeable in specific areas, such as audio production in his case, potentially revolutionizing how expertise is shared and consumed online.

Mike’s vision extends beyond mere functionality; he considers the broader implications of such technology on the creator economy. He speculates that as AI becomes more capable of producing content that rivals human output, the value of unique human contributions will increase. This could lead to a new kind of content economy where authenticity and personal touch become the premium commodities, as audiences may gravitate towards creators who can offer genuine interactions and original insights, setting them apart from AI-generated content.

The Bottom Line: How AI is Shaping the Future of Content

As we navigate through the labyrinth of AI’s potential in content creation, the insights from Mike Russell illuminate both the vast possibilities and the cautionary steps necessary for maximizing this technology effectively. As AI continues to permeate various aspects of content creation, from ideation and design to audience analytics and video editing, it offers creators powerful tools to enhance their work. However, Mike’s experience underscores the importance of balancing these tools with human creativity and intuition to ensure that content remains engaging, authentic, and legally compliant.

Mike’s perspective is particularly valuable in highlighting the dual nature of AI as both an enabler and a disruptor. While AI can streamline processes and introduce efficiencies, it also presents new challenges that require careful navigation, especially in terms of commercial and legal safety. Creators are encouraged to use AI as an assistant rather than a replacement, maintaining a strong hand in their creative output to preserve the human element that resonates with audiences.

Looking ahead, Mike envisions a future where AI’s role in content creation continues to evolve, making some aspects of manual creation optional and allowing creators to focus on higher-level creative processes. This evolving landscape suggests a future where the intersection of technology and human creativity will continue to redefine the content creation process, offering both exciting opportunities and new challenges.

As we embrace these technologies, let’s move forward with a mindful approach that leverages AI’s strengths while safeguarding our unique creative voices against the potential pitfalls of this digital frontier.

Resources:

SHOW TRANSCRIPT

This transcript is automatically generated by Descript.  Any errors or omissions are unintentional.

Jeff Sieh: [00:00:00] Hello, folks. Welcome to Social Media News Live. I’m Jeff Sieh, and you’re not.

Conor Brown: And I’m Connor Brown. And this is the show that keeps you up to date on what’s happening in the world of social media and more.

Jeff Sieh: Have you ever found yourself wondering how AI can take your content creation to the next level? This Are you curious about the ways AI can streamline your creative process and boost your productivity? Or maybe you’re looking to harness the power of AI to grow your YouTube channel and reach a wider audience.

If those questions resonate with you, well then you’re in for a treat today. We’re excited to welcome a guest who has mastered the art of blending AI with creativity. He’s been using AI to elevate content and grow his business. Mike Russell will be unveiling his experiences, his wisdom, and his best strategies for leveraging AI in your creative endeavors.

So sit back, clear your schedule, clear your mind, and get ready for this week’s episode of Social Media News Live. Mike, how are you doing today, my friend?

Mike Russell: Hey, Jeff, man, I’m really cool. And it’s actually an honor to be here. You’re such a gracious and [00:01:00] amazing and energetic host. So to be with yourself and Connor, I know this is going to be incredible.

Conor Brown: Awesome. And we’re so excited to have Mike in, but if you don’t know who Mike is, You should. Mike Russell has over 25 years of experience in radio, podcasting, and audio and video production. And Mike works with some top audio brands like Beyerdynamics, Mackie, Adobe, and more. He’s also an AI educator and the founder of Creator Magic, where he blends AI with Creativity.

As Creative Director at Music Radio Creative, Mike collaborates with a team that assists clients with their audio needs worldwide. We’re so excited to talk creativity, AI, so much stuff. So thanks again, Mike, for being here.

Jeff Sieh: This is gonna, we’re talking about creativity, but I have to give a shout out to our show sponsors who helped me be creative. It’s our friends over at Ecamm. You can find out more about them by going over to Ecamm. com forward slash Jeff. If you use my code JEFF15, you can save 15 percent on your first purchase.

And by the way, Creator Camp [00:02:00] is coming up. I know they’re sold out. There’s no more tickets, but You still have a chance. You can go to jeffsieh. com forward slash win and enter to win a ticket to Ecamm’s Creator Camp. Once again, that’s jeffsieh. com forward slash win. That’s coming up in September. Connor and I are going to be there, and it’s going to be a heck of a lot of fun.

It was amazing last year. All right, so, what were you going to say, Connor?

Conor Brown: It’s October, not September, because if it’s September, then I have to change everything.

Jeff Sieh: sorry. It is it. Yes, you’re right. I have so many September is Momentum. October is Ecamm. Ah, so many things. All right, so we’re gonna jump right into this because, um, I’ve got so many questions for Mike and he mentioned, we mentioned in the intro that he’s got this amazing new YouTube channel called Creator Magic.

As soon as this show’s over, you guys need to go over there, hit subscribe because he is giving great advice for creators on how to use AI in their business and in their workflow, so it’s, it’s amazing. He also has a podcast about that. Uh, it’s right there. There’s another link to his, the podcast. Other channel that’s the podcast.

So you need to sign up for both of those because [00:03:00] it’s really, really good. All right. Um, Mike, I want to ask you really, really quickly. you started this new YouTube channel, Creator Magic. Um, how has AI influenced that growth of that YouTube channel? Because I think it’s really fascinating. One, the way you’re doing it using thumbnails on it, but how are some of the, like the ideations that you’re doing to come up with video ideas?

Mike Russell: Yeah, it’s a great opening question, Jeff, and AI, in answer to your question, has helped me every step of the way with this new channel. So, uh, as you mentioned, I’m a bit of an old school YouTuber. I don’t like the word old, actually. Young and experienced YouTuber, I don’t know. Anyway, I’ve been at it since 2010, uh, which is insane to think about.

That’s when I uploaded my first video. So I can’t put it all down to AI. I can’t say, well, it was all AI. You know, I’ve got, YouTube Channel and AI has made it explode. Um, there’s been a lot of things that have, uh, contributed to that. Uh, first and foremost is just knowing YouTube, knowing how to put together a good video.

So I’d say to any creator listening to or watching this show right [00:04:00] now, if you’re wondering if there’s like a magic secret to how to make a big YouTube channel, Just keep making stuff and keep learning because it’s you learning as a human being and picking up these tips that will really help. With that said, when I went to create this new channel at the start of this year, I sat down and I had AI by my side all the way through.

And where AI really helped me was to fill in my information gaps and my knowledge gaps that I didn’t have. So I was determined to start a new channel about AI and help content creators learn these tools. And obviously, as you’ve mentioned earlier, my speciality is audio. So in terms of doing the audio, that was absolutely fine.

You’ve also mentioned thumbnails. So I now have a specific process in place where I’ll generate images, and we can talk in more detail about this later if you like, and then I’ll do a face swap. I’ll put me onto the thumbnail. I’ll put some text, just a couple of words on the thumbnail. I’ll come up with a catchy title.

And that will really help me get a good visual brand. [00:05:00] I also helped, uh, with AI in the early stages where I actually created the logo, 100 percent with AI. I know we were talking about that before we, we hit go on this show, how you can sort of ideate through logos with different AI tools. So I did that, created the brand.

Uh, I had in mind that I want to kind of magical wizardy kind of. Theme to it. So hence create a magic and the thumbnails of me dressed up as a wizard and magic tools and things like that. So I guess it’s mixing or blending human what it is to be human with ai. Come in with some great ideas and direct ai because at the moment.

As we all know, it’s, it’s not doing our jobs for us as creators. It’s, it’s assisting us. It’s a brilliant assistant. It can make us better. Um, sometimes it can give us the wrong advice, which is why at the moment, I think it’s good to seek an expert if you’re not sure about something. Uh, but that’s just a brief touch on, on how it maybe has helped me so far.

Jeff Sieh: and so real quick, uh, Connor, before your question, I want to, um, so George says, uh, that he, uh, went [00:06:00] over already and subscribed to Mike’s YouTube channel. Way to go. Everyone should do that. Um, and Gary says he’s the man for audio tips, George. So just to let you know, and he said, uh, he’s, and Gary’s a big fan of your live Friday show that you used to do on Fridays where you would do free jingles.

I remember those. Those were amazing. Yes.

Mike Russell: Yeah.

Jeff Sieh: So, um, but so you got your fan club here, uh, but I just wanted to say that, uh, you know, using the AI and your creativity together is what we’re going to really dive in in the second segment today. So, Connor, go ahead with your question, yeah.

Conor Brown: I first love, um, uh, I love Chris. Chris stole my joke. I was going to say, you’re not an old school YouTuber. You’re a classic

Jeff Sieh: That’s right, that’s right.

Conor Brown: classic or vintage better than, than old. So I love the, the idea, but going back to what you said, I think it’s really important. It’s creating a lot of stuff and just seeing what sticks and just create, create, create, learn and learn, learn, especially when you’re on YouTube.

And it always comes back to the thing of it has to be good content. AI is not going to take a awful video and make it [00:07:00] incredible and everyone’s going to see it. You have to come with your personality. You have to come with your story. And if it’s something that you’re teaching someone how to do something, they have to get something out of the video for it to be a success.

But AI helps in every step of the way. So. Can you get a little more specific about when you’re creating a video for YouTube, the actual like AI tools that you can’t live without now. So some AI tools and technologies that have been really impactful in optimizing this new YouTube channel that you’ve started.

Nice.

Mike Russell: and exciting tools this week. Okay, let’s start with that. How do I curate what I’m going to talk about? So, I will go to X and I’ll use, I’m a, I’m a signed up member of X, so I’ll use Grok and I’ll say to Grok, um, because Grok is very of the moment, I’ll say to it, tell me, you know, what’s new in today’s world.

AI tools this week. Sometimes I’ll say specifically AI tools that [00:08:00] creators can use, and it will summarize. It’s absolutely brilliant. I’ll go one, two, three, four, five, and I’ll be like, brilliant. Okay. I know about that one. Don’t know about that one. Need to check it out. And it also references. So it will source, um, I want to call them tweets, but X posts now, X, the X posts where it got the info from.

So I’ll check out those posts and usually that will send me down a rabbit hole. You’ve got to like doing what you do on YouTube as well. I should add that. If I didn’t like looking at AI tools, I’d find this incredibly boring. But I am just a tech nerd. I love technology. I love new things. So I’d kind of be doing this anyway, but now it’s just I get to curate it and talk about it.

X is where I’ll start with Grok. That really helps me. Uh, I’ll also use a website. Uh, I’m sure many of you will know called Product Hunt, and I will just scour and see what’s been, you know, peaking up and what’s new this week. All the new announcements tend to come there. And then if I get really stuck as to like what to do or how to place things, sometimes it’s obvious to me, there’s a new tool and I’m like, well, everyone will want to know about that and I’ll do it.

And [00:09:00] I’ll put that But sometimes it’s a little harder. So I’ll create a list from everything I’ve researched, and then I’ll put it into, say, ChatGPT, and I’ll say, I want you to go through one by one and tell me, you know, uh, what your thoughts are on this tool, how it might be useful to creators, and give it a rating out of 10 as to how valuable it would be.

And that’s it. I will preface this by saying it doesn’t always nail it and don’t trust it because it might hallucinate or say incorrect information, but at least it can give me a guideline. I can see, Oh, look, these are nines and these, this one is a 10 and yeah, I can see why it would do that. So it helps me to.

rather than scouring through a list of 100 tools to quickly determine, right, these are the main tools I should be looking at this week. Then I will go ahead and record a video. So once I’ve got everything curated and listed, and I use my Elgato prompter, which is why I’m looking directly at you as I talk to you, sorry, going out of focus at the same time.

So that’s really handy. And I’ll throw up bullet [00:10:00] points on the Elgato prompter. So never put a full script there, because it’s really mechanical if I read.

I’ll just, you know, bullet points on each tool that I want to get across and then really Jeff and Connor, I just, I just have fun. I just, I literally, the way I record my videos now for this weekly roundup, which is, I would say that’s the thing that’s growing my channel fastest at the moment is that weekly cornerstone content that everyone has now come to expect.

I will load up, uh, I actually use OBS to record. It’s perfectly sufficient for my needs and I record my cameras and my screen share. I hit start recording and then I forget I’m recording and I just mess about with AI tools. Usually takes me about an hour, hour and a half to do that, um, because of all the, you know, the production and things break and things don’t work.

In fact, I could probably make a compilation of the AI tools you must miss this week. Um, it’s all the rubbish I’ve been through. That’s the often, and I know as YouTubers and creators, you’ll understand that, uh, often we have to sift through the [00:11:00] rubbish to get to the good stuff. And that’s kind of, at the moment, AI can do a better job.

That’s kind of our job. You know, we’re all curating or helping people understand something, and they come to us because we help them to find the best of that thing.

Jeff Sieh: one of the things that I’ve never heard before, I’ve never had anybody on the show talk about Grok and using that. But it makes sense because that is the news and a lot of times, you know, Twitter has a lot of the updated stuff. That’s really fascinating. I’m gonna have to test that because I have not even gone down the Grok, uh, hole in the ground.

So,

Mike Russell: So incredible. It’s so of the moment. You can ask it anything and anything, even emerging news stories that are unfolding. You can say, what’s your take on this? And it will, it will have a take. I’ve been using it during the, we’ve got the Euro soccer is I think you’d call it over here. We’ve obviously got the Olympics.

If something is changing in like a sporting event, I can say, what does this mean? Like I can’t remember what it was. I was watching something like. I think I was watching the uh, it was one of the Olympic [00:12:00] events and I didn’t understand, it might have been rowing, I didn’t understand how they all got through and eventually got the medal.

So I said what does this mean that this has happened and it was able to sort of unfold it for me.

Jeff Sieh: That is really cool. Uh, Chris says, I’d like to subscribe to your AI rubbish channel, please. So, that’s funny. So, yeah. Um, yeah, that’s hilarious. Uh, and then, like, Gary, so, we got some people chiming in, like, how they using it, like, Tech Troublemaker saying, I use, uh, Chat 15 nearly every day for a lot of stuff I do on YouTube.

Uh, Gary Stockton says, yeah, me too. The models are so incredible now, just about for any task. And, um, one of the things that I want to talk, ask you a question about, because, uh, one of your podcast episodes, there was a lady, and I forget her name, but she was a big podcaster, and she was using AI for a lot of stuff, and she was actually running it through How this, this, I think it was open source something, but it was giving her audience sentiment on her, how her podcast was coming through.

And I’m like, Oh my gosh, that’s amazing. Um, so do you use AI to analyze like audience engagement and feedback on your videos? Are there any tools that you use actually with [00:13:00] the stats of the videos to kind of inform you on what you need to do next for your next video?

Mike Russell: That’s a really cool question. So I think that was, uh, Steph Fuccio, uh, who was on the show. She’s the AI podcast consultant and she was using a tool from Assembly AI. I think it. That’s their Universal One transcription model. It’s pretty cool. Uh, most, most people or most benchmarks, sorry, I should say online, are suggesting that it’s even slightly better than OpenAI’s Whisper.

So it’s a pretty good transcription tool. And like you mentioned, Jeff, it’s got a sentiment analysis. So I can say this percentage is positive, this percentage is negative, et cetera. Um, I haven’t played with it extremely in detail myself, but I know that’s very interesting. You were specifically asking me though, You know, do I use any tools to help me see the audience, uh, sentiment and how they’re doing?

So I use the, you know, the standard stuff that you’d expect, like I go into YouTube Studio and I look through the analytics and I try to see, you know, where people are coming from. I’m particularly [00:14:00] interested in, um, where people are paying attention. And, um, okay, so AI can assist in this. I find it more helpful in long form content.

Yes, it’s, it’s great with my You know, my AI channel where I’m doing a tool or I’m doing a weekly roundup. Often I’ll see the, the peaks at, you know, when I’m talking about a new tool. So it’ll go up, down, down. But when I’m doing a podcast like this, like maybe 20 minutes or half an hour long, I like to go back and look at where the attention points are.

And this is really interesting. Cause I’ll find the points where the audience were like, That was an interesting point. Now, this is where I’ll use AI. So I’ll, I’ll download a transcript of that podcast episode we created. Uh, I try and have it speaker labeled as well. Actually, Premiere Pro will do that for me.

So, that’s the editor I use to edit the video and audio. And it will give me a transcript with speaker labels. And then, after a while, I will look back through at the YouTube studio and I’ll find those attention peaks. And then I’ll mark them out in the [00:15:00] transcript. And then I’ll copy and paste them into something like ChatGPT or Claude.

And I’ll say, right, these are three really important points from a podcast. Like these really grabbed attention, create me three very short headlines or whatever, or promotional materials based around this. So I can have social media posts, etc. Ways to promote what was said on the show. Um, but more importantly, and something I’ve really found helpful for optimization.

is thumbnail testing. So on YouTube, as you know, I think most people now, if they have a verified account, they can upload three different thumbnails and test them. So I will then, uh, for the podcast, I put text, I put a quote on there. So I will then try and optimize those quotes based on attention, paid. So obviously I have to upload a thumbnail initially, but then later on, I can go back and optimize those thumbnails and split test them to see what exactly is working. And that’s pretty incredible. Uh, and YouTube doesn’t quite optimize for click through rate. [00:16:00] So it’s not looking to get you the most clicks, uh, but it’s looking to get you the most watch time.

So it’s. Your thumbnail matches up with the audience’s expectation and you’ve got that beautiful, perfect storm and people hang around for longer. And I found actually split testing thumbnails brings me significantly different watch time. And it’s very, very interesting to, uh, to, to play around with that.

Jeff Sieh: And I think I mentioned, you mentioned on Mike Stelzner’s podcast that you’re using Mr. Beast’s new analytics tool to kind of help find some of the, you know, the right thumbnails and quotes and stuff that go down. Okay. I want to double check that. So, cool.

Conor Brown: That’s very interesting. You know, we, um, Jeff and I, uh, I participate in, Jeff puts on a, a paid mastermind and we met yesterday, but I shared a, a link with all the gents in there to this new website I found called There’s an AI for That, and I was just scrolling through it and, you know, it’s separated by, oh, it’s, there’s an AI for audio, for video, you can type in and, and search and all that stuff.

And as I was scrolling, it was like, okay. And then last week, here are the ones [00:17:00] like, oh my God, I scrolled through like, All of those are some of the top 200 AIs that were brand new just this week, so it’s always changing. There’s always something new. I like, Mike, how you said some of them are crap, Some of them are great.

You gotta test to find out, right? All of that. and I think we’ve talked a lot about how AI can help in the Creation, the Summerization, the You Getting key points from what we’ve done. But when it comes to the actual editing and the production process, Mike, are there any AI tools that you really, really find useful when it comes to actual production and then editing once you have that, uh, uh, produced out?

Mike Russell: I know a lot of people like to use, uh, tools, uh, such as, I think there’s a tool called Glean, uh, which one of the guests on my show mentioned, and it can automate the editing of, uh, the video and audio in Premiere Pro, for instance, I think that’s it. G L E A N, Glean AI. Uh, I haven’t used that myself.

And actually, when I’m actually editing my show, I tend to stay away from AI. And I know this is going to sound counterintuitive and [00:18:00] strange. Obviously, I’m experienced in audio and video editing. So I, I kind of know what I’m doing. And I’ve played around with tools similar to that, not that one specifically, but these tools that do auto cuts and auto edits and stuff like that.

Um, even, you know, a tool I use all the time and I advocate for is Riverside FM. I will record, uh, podcasts in there and there’s a load of AI tools in there to like, cut out silences and filler words and stuff like that. But I personally, as an editor who’s basically done it all my life, as you mentioned at the start of the show, working in radio, I’m very specific and I know like 90 percent of the audience might not notice that, but I’m just like, I watch AI do cuts and I’m like, no, that’s, I don’t like that.

That’s not cool. That’s not good. I don’t like that. For me, for the 10, 5%, it’s like, it just doesn’t, it’s not right. And I actually think, yes, it might take me a little bit longer, but I’m, as we all are, when we do something again and again, once I get on that kind of flywheel of doing it every week, I just sit there in Premiere and I don’t even necessarily have to play the words back.

I just skip [00:19:00] through and I’m like, right, that’s a breath, this, that’s that, that’s this, cut that, cut that. I had been using, in the earlier days, AI. to help me as an editor to cut stuff out of a podcast. So I don’t, I don’t do a show like this one where it’s live. Everything that’s said just goes out there. I will actually chop, chop, chop, chop, chop to try and put, squeeze all the value in.

And I have asked AI, I’ve put transcripts of the, like, we might go for 30 minutes and I might say, help me cut this down to just like the important stuff. And if we go on a tangent, tell me about it. And it kept saying that all the, all the bits I said in between the guests, you know, the filler bits where you give the.

To think, and you say, you tell a story, it was saying, that’s rubbish, cut that out. And I’m like, hang on, I don’t like this AI. It’s telling me just cut my stuff out. So this is maybe it’s right. Actually, it’s probably right. It’s like, Mike, you’re talking nonsense. Stop, just shut up and let your guests speak.

Jeff Sieh: Well, Chris Stone and I have this talk a lot because we’ve tried a lot of the tools and, you know, there’s a, there’s a son, like [00:20:00] Opus and Mimeo and all this stuff. I still don’t think they’re there yet, like as an editor, like, um, I think you and I both, Mike, I started, I think I had Premiere 2, like I’ve been doing it, video editing for a long time because I’m old.

And, um, one of the things is I like to use like Descript as a new, their AI underlord, which I really like. And it gets me most of the way there. Like it, it finds those clips, it’s not perfect and I need to extend them or trim them or whatever, but it still saves time as a tool. Now the creative stuff is me picking and changing and moving stuff around for the shorts or whatever.

So I think that’s helpful. But it’s still, and I don’t think it’ll, it probably won’t ever be one button click and you’ve got 20 shorts done out of your show. I just don’t think that’s, If it, if it happens, then, um, I still think having your, your finger on it is going to be what’s super important to grow your audience.

But that being said, I want to kind of segue into that because I want to talk about AI and the creative [00:21:00] process. Yeah.

Mike Russell: I’d love to pick up on your comment on Opus. Uh, so I agree with you. I think I, so I’m a signed up member of Opus Clip and I think there’s a lot of work involved with putting those shorts together. Once it’s isolated, it said, right, here are your 20 clips. Like I’ve been in there. I’m like, actually, this is a job I could probably outsource.

Cause I’m going through, I’m like, right, cut that bit, cut that bit, like tidy that up and generate B roll. And then, oh no, that B roll looks weird. Let’s generate again. And I’m like, this is, you know, until it spits out 20 clips and you can just go. Schedule,

Jeff Sieh: Right.

Mike Russell: it’s still taking time, isn’t it?

Jeff Sieh: Because I even, I go the step further into script, I get those clips, and then, you know, you can export the scripts into Premiere. So if I want to really finesse them, I can open it as an XML file, the Premiere thing, and I have all my tracks in it. So I’m actually able to split it down, but then the finessing still happens a lot of times in Premiere, I know Chris Stone does that.

Uh, Gary had a, a great comment, he says, There’s one thing in the script I find frustrating, I can’t seem to, I can’t just see the text. The tracks for my video in the multicam edit. Gary, let’s get on a call because I think I can [00:22:00] show you and help you out with some of this stuff. Uh, and I know you’re in my course, so we’ll, we’ll, we’ll, I’ll make a little video for that because I think I can help you out with that.

But anyway, yeah. It’s always, and the other thing, it’s always changing. Like, Opus is getting better from when I tried it, I’m sure. Uh, so is Descript, so anyway. Uh, and then Katie says she likes CapCut much better for Opus in creating better clips. So, yeah. You know, it’s just, there’s so much stuff everywhere, it’s so fun to nerd out about this stuff, but you’ve got to find the process for you.

Um, on this note of, you know, creativity, like, a lot of people are, and a lot of creators, and it was the same way when Canva came out. All the designers were in an uproar, this is wrong, you know, and the same way with AI, people are going, it’s killing creativity. I see it as a tool to help me be more creative.

And I want to kind of dive into this with you, because you know all the tools, Mike, you’re in it, your channel’s all about it. So, let’s talk about ways AI can assist creators in brainstorming and developing some new content [00:23:00] ideas. Kind of the blank page syndrome is like what I like to call it. So, what have you found helpful in that way?

Mike Russell: That’s a, that’s a really good question. And I think there’s, there’s quite a lot to unpack from it. Um, so yeah, blank page syndrome. Yeah, definitely. As creators, I think we all suffer from that moment where we just don’t know what to say. We don’t know what to talk about. Uh, so let me share from my own experience.

I mean, yeah, Even this new channel I have, sometimes I think, you know, what am I going to talk about this week? How am I going to make something that’s interesting and exciting? So it can be quite difficult sometimes to actually come up with those ideas. Um, one of the ways is just simply to go to your favorite language model, like ChatGPT or Claude, and say, you know, give me, Give me a few ideas based around my channel.

Feed in the information from the channel. I suppose that’s a very sort of basics way of doing it. Another way, if you’ve already got some content, is to feed in the content you’ve previously done as like custom instructions and say, all right, these are the kind of videos I’ve done. These are my [00:24:00] last 10 videos.

Give me an idea for the next five. But I do find that AI is still very fluffy and generic. It’s pulling from training data that is, you know, trained on various different sources, which may not be your voice and style. So I find stylistically it’s hard, uh, especially when it, you know, it’ll look at my channel and it’ll say, five mind blowing tools to enhance your creative flow, or, you know, seven game changers that everyone should use in their video production.

I’m like, I don’t talk like that. It’s like, I don’t see any game changers. And often they’re not game changers, they’re just good tools. So, um, so yes, that can help you get started. But again, I’m going to stress like, and I think it’s really important, the, the human should come into the process, uh, with, with something.

You can’t just say, I run an AI channel, give me some ideas. It’s much better if you can be more nuanced than that and really tell it what you want. And, It’s like having the best intern you’ll ever have, you just give it the instructions and the more information you give it, the more you will get back [00:25:00] because AI is so good at mirroring right back to you what you’ve asked for.

Uh, so the more you ask for and the more detailed you can be, the better. And I think one of the greatest things with context windows now they’re so big, you can feed in a lot of data and get more out. So in the early days, you know, you’d, you’d feed in an article and it’d be like, You know, out of tokens, can’t, can’t do this, too large.

But now you could probably feed in the transcript of your last five or ten videos and say, this is the way I talk, this is what I’ve done the last ten videos, now give me some ideas and speak in the style of these transcripts. So that’s how I would dial it in further.

Jeff Sieh: Yeah, I think that’s really important. And the other thing is, before we, Connor, you ask your question, is that I think a lot of times, and I was guilty of this. When I first started, I was using AI for everything. I delved and, you know, dived into all those words that AI used, and I was using it for my email newsletter.

Well, now I don’t do any of that. I want the newsletter to be me, and I want people to know it’s [00:26:00] not AI. So my newsletters are a little wacky, so people know that they’re not AI at all. Um, and I think that you have to sometimes zig when you can zag. Like, yes, use AI to speed up your ideation and help you come up maybe with a logo and all that stuff.

But when you’re making a direct contact or you’re trying to engage with your audience, it’s gotta be you. Like, I don’t see Mike ever just going and doing an AI avatar of you that you just type, and it automatically You know, has your mouth moved realistically? I don’t think I’ll ever see that on your channel.

Maybe I’m wrong. Uh,

Mike Russell: Zuckerberg wants to make that happen. It’s like he said, he wants to make all the creators have their own avatars trained on their data. So who knows?

Jeff Sieh: maybe, but my thing is, is like, that’s why I love this live show is because I will stumble over my words. I will say things wrong, but people, AI can’t copy this right now. Right. They can’t, can’t. And that’s what I like about live video. And I think that touch point is super, super important,

Conor Brown: Well, I think, you know, we talked about a great thing and Mikey brought up a great thing about AI being a great intern, but then [00:27:00] at the same time, other AI tools where you feel like, well, this has given me so much, I almost need to outsource this using this AI. Pretty frequent question that’s come up is, when you look at external tools that you know you’re using is not enough.

Then you’re going to start separating these, trying to start re orienting your processes when looking at internal tools that you’re using. I think, you know, Bye. is giving me more useful time management right then taking away time from things that I actually should be focusing on because it’s very easy to get distracted.

So when it comes to like just the audio and the video production, when you have a creator and a solo creator who’s just starting their YouTube channel or their podcast or, or something like that, how do you think AI can enhance their production. I guess from the creation of an idea to actually recording something, especially when they have super, super limited resources to start with.

Mike Russell: I like what you said there, Conor, about being [00:28:00] distracted by all the shiny tools. It’s so easy to do so. Um, yeah, but if a creator wants to get started with AI, you know, from start to finish making something great. Uh, I would say. Go ahead, pull in, pull in the sources of what you’re going to talk about.

So, for instance, for me, if I wanted to talk about a microphone, I might go to three or four different websites. Actually, I might not even do that. In the past, I would have gone to three or four different websites, you know, taken a look at what was said, manually make notes about the microphone, like, pull out the interesting parts, the unique parts of that microphone, and make my own list.

In the, In the past, I might have gone and copied and pasted to chat GPT and said, tell me the interesting stuff. Uh, another tool that’s really good if you’re a creator and you want to get the lowdown on things, I use a lot of Perplexity. I’m sure you guys do more so than Google now for doing research. In fact, it’s so good that they’ve been through, I think, a few legal, uh, cases recently about, like, spreading stuff from the internet.

Um, but now they’re like, but don’t worry, we’re going to pay people who [00:29:00] make this stuff. So we’ll see where that ends up. Um, But perplexity is just incredible. So I’d say actually scrap everything I’ve said about ChatGPT and custom instructions. Go to perplexity. Where perplexity really shines is if you want something specific.

So you say, I want to know everything about the Neumann U87 microphone, and then it will tell you everything about that. It will Pull in from various different sources, and then you’ll be like, okay, now tell me the best features about this microphone for a content creator. And it’ll go and further refine it.

And, uh, yeah, I mean, it’s, it’s, that is a really cool source of inspiration. And often I’ll end up, once it’s given me some, some basics, I’ll then my last prompter, it’ll be now make that into four or five good bullet points, uh, you know, that I can put up on my prompter and it will give me the info that I need.

And that has helped me tremendously. Especially when I’m talking about tools and I want accurate information. And obviously we know ChatGPT Cord can hallucinate, some of the other models can. [00:30:00] But perplexity generally tries to take its information and, and site, site where it’s got the source from. So you can be a little more confident with that.

But it’s just an incredible tool, perplexity. Uh, it’s changing the way, just, just everything now. I just have the app on my phone. And if I have any question, I just straight away, I go there and I say, Help me with that. So yeah, this is, this is a good way to get started. And then going into putting out the content, I’d say, you know, hit record, get it all laid down.

I mean, like, like you say, Jeff, you know, get it into something like Descript, which has its overlord. And then you can stop recording and correct me if I’m wrong, but I think overlord will go through, it’ll tidy things up. If you do a retake, it might, might be able to cut those retakes out and give you a nice, clean, polished thing.

So you might’ve been recording for 30 minutes and you might get something, That’s down to four minutes once you’re done with it. Uh, and then I think again, Overlord can do things like generate, you know, shorts with, uh, captions that animate and things like that. You can output all of that stuff. Um, and I think, yeah, we’ll [00:31:00] eventually we’ll have something like that as a bigger.

AGI assistant that will just assist us in the future.

Jeff Sieh: Yeah, and it’s actually describes Underlord because Overlord’s too scary.

Mike Russell: That’s the one.

Jeff Sieh: That’s, that’s what they call the Underlord. Yeah, yeah. Because it’s Overlord. Overlord is too dang scary. Um, so, Gary says, um, he likes AI for drafting an outline for an article or video, and then we fill in, uh, the rest in our own voice.

And, yeah, I want to talk, when we can talk about the future, I want to bring up Gary’s comment. I’ll bring this up, Gary, when we, when we kind of go into the future of AI. But I wanted, on your point about drafting an outline, is there any, Thing that you can think of, Mike, where creators really need to watch out with their using AI.

I mean, I think we’ve kind of talked about, don’t just copy and paste because that look, you’re going to have to rewrite some of this stuff, but are there any other pitfalls or challenges that you see, or you want to warn creators about not to use AI or how not to use it? [00:32:00] Right.

Mike Russell: Let’s talk about fact checking. Let’s talk about commercially safe content. Those are three pillars I think creators should watch out for. So everyone knows that AI can hallucinate. So, you know, don’t go to, I think it was Google that had the big, uh, outrage when they said you should eat rocks for your health and put glue.

You know, it can do terrible things and if you repeat word for word, then, you know, you’re in trouble. So, so fact check stuff, you know, chat GPT says something, you know, use Google and check it and then go to perplexity and double check it, triple check it. Uh, you know, there are ways these, AI creators are trying to fix that.

I think Google on their Gemini models. Did have maybe still does have a fact check thing where it’ll highlight things in red It thinks might be false or need checking and highlight things in green where it’s like yeah, this is definitely good information So they’re trying to fix the hallucination problem, but it’s um It’s definitely not an easy one to do.

So fact check as much as you can. [00:33:00] And then as creators, I think we should be very aware of using generated content that’s commercially safe. Now, I know this is super important to like agencies and, uh, you know, fortune 500 companies that absolutely cannot afford to put anything out there that, you know, might be questionable in terms of legality.

So, uh, Creators may care a little less about that, especially when we’re at the start of our journeys, you know, and we’re just kind of trying to get stuff. And, you know, back in the day, we might’ve thought it was okay to go to the Google image search and type in, oh, that image looks good. I’ll put it in my blog post.

And then you get like a notification from like, I don’t know, Reuters or the associated press. That’s ours. Don’t use it.

Jeff Sieh: too. I’ve seen some people get spanked with that stuff. It’s yeah. Yeah, not good.

Mike Russell: So that’s not good. And then, yeah, in terms of commercially safe, you know, Mid journey, you know, what was it trained on? I know we all love to use it. Um, could it be a problem in the future? Maybe it won’t be, but maybe it will. How much level of risk are you [00:34:00] open to there? Um, Adobe are really doing their best to do things, well, they are doing things the right way.

They use the word commercially safe everywhere you see the word Firefly. So you can be sort of guaranteed that this is going to be good. It’s backed by Adobe. I know other companies like OpenAI have said. You know, if anything DALI produces is legally questioned, we’ll back you up. I think they said that last year in their November conference.

So I guess that still stands. But again, double check these things. Suno and Udio, I think we briefly touched on, might’ve touched on before we hit record, actually, but the audio music generators, really, really good. Incredible technology, clearly trained on copyright material and going through, you know, um, legal cases right now, I think with Warner, with Sony, with Universal.

We’ll see where that ends up. You know, I, my school of thought on this is that I think these big companies that are doing these projects, you know, they’re trying to do something really impressive so that we all go, wow. And then they’re asking for forgiveness [00:35:00] and let’s see what the courts decide on that.

But

Jeff Sieh: And, and I think the music, I think the music companies are upset because they’re not getting paid. I think if they got paid for being trained, they’d be fine with it. Really.

Mike Russell: there’s so much to argue about here, isn’t there? Because this is how we learn as humans. You know, we listen to, you know, something by Eric Clapton, and then we go, I want to, I want to play that on the guitar too. And then we, we listen to his song over and over again, and then we get the idea and then we start playing.

So we are training each other, you know, like that. So we’ll see where that goes. But like, yeah, I, I think, There’s, there’s two different ways of doing this, aren’t there? So there’s the UDEO and the SUDO way, and then there’s Stability AI, which have got stable audio, and they’ve been, again, very clear. You know, we’ve trained this on, you know, a licensed library, uh, which you can use, and all the other stuff.

I think they did a similar sound effects generator as well, and they said everything we fed into this is Creative Commons Zero, which means you can use and [00:36:00] remix without the need for attribution. So there is really clear as well stuff like that, but. In all honesty, if you use Stable Audio, uh, or you use Udio, you know, Stable Audio is comparatively rubbish compared to

Jeff Sieh: right, right. It really is.

Mike Russell: So, you know, what are you going to do? You know, when I preface the word rubbish, it’s great for background music on YouTube videos. You know, it’s great for like stock music and stuff like that. If you want to create a hit song or something really insane that people are going to love, Udio and Suno is the place for that, isn’t it?

Jeff Sieh: Yeah, it’s, it is. George even said it’s amazing. I, I found that Udio is, sounds better to me than Suno, but it’s, it’s an arms race right now, uh, to play with, but they give you free credit. So if you want to test it out, go there and, and go. Chris, Chris is, he, he, he’s, it’s hard to get him to tell his opinion, but Suno is trash is what he says.

So

Mike Russell: crazy thing is, Jeff and Connor, about UDIO. So I’ve used this quite a [00:37:00] bit myself, and I’ve also paid for the paid account. They have an audio upload feature, and I know you guys are aware, but my background is from an audio production company, make jingles and things like that. So we’ve got a lot of stock music that belongs to us.

So it’s our copyright. Um, and a lot of them are 30 second pieces, which can be like. sung over. So traditionally, you know, a client would come to us, they’d ask for a 30 second piece of music, and at the end, maybe the name of their podcast sung along with their tagline, right? So now what you can do is you can take the first 15 seconds of that music track, upload it to Udio, type in the lyrics, and it will throw you back that jingle with a professional singer singing.

And now here’s the question, because it’s our copyright, so that’s our music, and because vocals are slightly different to music in terms of copyright laws, Where do we land with that?

Jeff Sieh: Oh gosh. Yeah. It, uh,

Mike Russell: With the voice, you know, and generative voice, you know, copywriting of the voice is [00:38:00] extremely hard. You know, we, we’ve seen Scarlett Johansson go through the whole case with OpenAI and Sam Altman, you know, and, and she won’t be the last, you know, a lot of the famous voice artists, you know, are able to fight their battle.

But I know personally, Some really well known voiceover artists in the trade who have worked for years. They’ve been on all the famous TV shows and Saturday night specials and stuff. So you wouldn’t know them if you saw them on the street, but if, if they opened their mouth, you’d be like, Oh, it’s like people like that are like, I, you know, There has been a lot of, um, what’s the word for it?

Kind of apathy in the voiceover industry recently from my, my perspective. Um, because people are just like, well, you know, it’s time to find a new job, you know? So there is kind of a sad side to this AI where people are thinking that.

Jeff Sieh: we had Jodi Krangel on a couple weeks ago and she’s the voiceover artist and we were talking about that a little bit. One of the things too is like with the last kind of thing with Udio and Suno is [00:39:00] now I just saw last week or I think maybe it was yesterday that you can now download stems from your music.

So we could actually have a live singer singing to an AI background now because of the ability to do that which blows my mind. Like, this stuff is moving at the speed of light. It is just crazy. So, on that note, real quick, I want to Oh,

go ahead.

Mike Russell: is that yes, it’s great that UDO can do that. Um, but there are local models you can run on your own computer that can do stem separation. So many companies have done this. Audacity created plugins that are PC only that do that. Logic in their latest 11 update have a stem separator baked in.

And anyone who purchased Logic 10 for 199, Got access to that for free. And there’s another open source project called Ultimate Vocal Remover that works on every platform. You download it, you pull the ethernet cable out or switch your Wi Fi off and it will do the stem separation on device. It’s nuts.

Jeff Sieh: Yeah. It’s [00:40:00] just so crazy. Anyway, I want to give a shout out to somebody who’s always also going and innovating. It’s our friends at Ecamm. They sponsored the show. Thank you for them for doing that. Ecamm. com forward slash Jeff Sieh. Fifteen percent on your, uh, purchase to use the code JEFF15. They’re amazing.

Uh, once again, I mentioned that, uh, their camp is coming up. They are sold out, but you can have a chance to win at jeffsieh. com forward slash win. Go there and enter, uh, to have a chance to win a free ticket. So let’s talk. We’ve kind of dived into it already, but the future of AI. We talked about the music stuff.

Um, what do you, put on your, your little wizard cap there, uh, Mike, and, uh, tell us where you see the role of AI evolving in the content creation landscape in, let’s say, the next year.

Mike Russell: So the next year is going to be interesting. So obviously, uh, as, as we’ve all seen, it’s incredibly difficult to predict the future. You know, even the high level CEOs like, uh, Sam Altman, you know, like I can’t just can’t tell you what’s coming because I just don’t know kind of [00:41:00] thing. I’m sure like people like that do have a good idea.

And already are in talks as to where this is all going. Uh, in the next year, things are just going to get better. So, um, I think again, to reference Sam Altman, he’s always said, when you build something on OpenAI’s API, you know, don’t try and You know, don’t try and do something we could do, essentially.

Just build something interesting and assume that the models are only going to get better. And if you work with that in mind, you know, what was really difficult in, you know, one case may become easy in the next. in the future. So everything is going to get easier, is going to get better and more dialed in.

And I think ultimately beyond the year, hopefully beyond the year, otherwise I’ve really got to find a new job, is that we will be replaced as creators. And I want to say that now, on the record now, you know, this technology is here to make content creation optional. And I don’t want people to be as scared about that as it sounds.

Content creation will be [00:42:00] optional. Meaning that everyone who considers themselves a content creator today, and I think Mark Zuckerberg said at least 200 million people on his platform consider themselves content creators, so that’s a lot of people. Uh, you know, We will filter. I think we will filter down and we will have the people that really want to do this still doing it.

Um, and I think that in the, you know, maybe the beyond one year future, we will have people creating AI avatars of themselves that can answer questions 24 7. And providing you can dial that in and say, you know, only talk about this subject. You know, I’ve got no problem with having an AI Mike Russell who can answer all of your audio questions.

So that’s what I’ve been known for. over a decade now. So I’d be happy to create a clone of me, uh, you know, with permissions. I, I don’t know how the models will work. Maybe the, you know, you’ll get a little bit of payment like Google AdSense or something for every question that’s answered and upvoted or whatever that is.

Um, but I think that’d be incredible. And then you can just say just AI [00:43:00] Mike can talk about, you know, audio. He can talk about video, but if someone asks how to make an omelet, you know, don’t answer. Cause I don’t do omelets.

Conor Brown: well, it’s funny because, you know, here in the States, Olympic coverage is through NBC, one of our channels, and Peacock, their streaming service, you can get AI Al Michaels, Al Michaels, a legendary American sports commentator, to recap the day’s events based on your needs. So you can say, A. I. Al Michaels, I Want Basketball, I Want Gymnastics, and I Want Swimming, and he’ll give you a recap in Al Michaels iconic voice.

Of course, I don’t think A. I. Al Michaels could do Do You Believe in Miracles? Yes, but he can tell you if Katie Ledecky won in the 1500 freestyle, which I’m sure she did. Um, So I think that there’s so much with, with all of that, but it also brings up a great point that, that Gary brought up, uh, uh, a couple of questions back.

He says, the downside to all of this is zero clicks in search results for your [00:44:00] niche is resistance to AI utile. And I think to piggyback off of that too, you know, we hear more things like. Whether it’s Google AI Overview, whether it’s, you know, Sam Altman and OpenAI saying we have AI, uh, uh, what’s it called?

SearchGPT? Right, searchGPT. So those things. So the futility of it all, I mean, you know, Mike, it’s, it’s, I think you would agree we, we have to embrace it rather than just cast it aside.

Mike Russell: Yeah, we definitely do. And we need to think, you know, um, where can we be as human beings? That’s still valuable. So definitely, it’s the end of text based, uh, content in a way, I suppose. Is that, is that okay to make that as a statement? Because like you’ve just said there, Connor, you know, we’ve got SearchGPT, we’ve got Google’s, uh, you know, Search Overview.

where it’ll give you a little box at the top and highlight the important piece. I mean, to a certain extent, you know, I would say, you know, live on YouTube, create video, because you get the real [00:45:00] human interaction, albeit over the internet. Um, but then Google also has something where it will skip just to the point in the video.

So if you want to know, you know, How to change an inner tube, you know, you’ll do that and it’ll be like, right, watch 1 minute 14 to 1 minute 28. That’s where the creator covers it rather than the whole five minute video. Hey guys, welcome to the show. Remember to like and subscribe, you know, it just gets the content.

So we’re going to be forced to be, uh, more specific, but I think people are also Still seek out that real human interaction and that knowledge that they are connecting with a real human. And again, you know, another one of my favorite quotes coming from Mark Zuckerberg is we are going to have potentially holographic versions of ourselves.

And imagine what it will feel like when we will be able to interact with people and we will be able to really, truly, genuinely feel like they’re in the room with us. And I don’t think we’re too far away from that happening. In fact, it would be incredible to have an experience where we’re doing this show now, and I really genuinely [00:46:00] felt that you guys were here in the room with me.

I thought that would be incredible. Uh, it’s just something that, you know, that’s why we go to conferences, isn’t it? Because at the moment you can’t really replace that experience. Oh,

Jeff Sieh: you’ve played with the Apple Vision Pro yet. I know it just went over to the UK. Um, but, so, I have one, and when you get yours, Mike, we need to connect and do some videos with that. Um, because once you try it on, you’ll get one. I know you, and you will. Um, and one of the things is, it’s, it’s still a little uncanny valley doing the, um, you know, the, the personas they have.

But the memories that you can take now, and the new phone’s gonna come out, and the, the, uh, the one right now can, can take the spatial videos now. There is nothing like seeing that because it feels like you’re there. It’s very Harry Potter pensive kind of thing. And you can’t, you can’t say it with words until you’ve actually strapped it on it and done it.

Like I have some videos of my dog that I’ve taken. I wish I would have had this when my kids were little, uh, to be able to record this stuff. So, what you’re saying, I think, it’s, it’s on its [00:47:00] way. Like, it’s coming. Like, they’re working on that stuff, so, I think it’s an exciting time. I do want to double down on that I think the, the more AI happens, and the more we don’t trust the written word, we know maybe it’s created for AI, that people are going to pay.

To be in Mike Russell’s group where he teaches me about audio, but it’s real. And I can ask him questions in real time and it’s, and it’s Mike. So I S I think that’s going to go up for creators. If they’re going to jump on that opportunity, that’s just my two cents. So, um, Connor, um, did you have a question?

Conor Brown: well, I’ll just also say like, there’s so much for better or worse, that we as humans, we appreciate others opinions that we trust. And I don’t know how necessarily people will appreciate the opinion of a third party like a Google like an open AI that is sourcing something from Google.

Who knows where right? So I don’t know if that’s what I always go back to as well. Like, if you’re able to still find the people and they can trust you, your opinion becomes valid and you’re also including facts and [00:48:00] reasoning and all that stuff to back up your opinion. And there’s so much coming, of course, whether it’s, you know.

What Jeff was talking about with the Vision Pro or all these new tools that happen every single day, but like, Mike, upcoming AI advancements, what are you just excited about? And let’s just say as a human being, when it comes to the AI advancements, what are you looking forward to?

Mike Russell: So much, Conor, so much. I mean, yeah, just that being in the room together thing. I think that’s gonna be incredible. The, uh, like Jeff, like you mentioned, the Apple Vision Pro, uh, MetaQuest, uh, actually when I was last in the US, I, I went into Best Buy and I put one on my head and I was like, this is nuts. And the whole, like the whole store melted away and I went into this.

virtual reality world. So this is definitely exciting. It’s only going to get more, uh, realistic. Um, it’ll also be succinct. So I think we we’ve spoken a lot on this show, uh, over the course of the last sort of hour or so about the need for experts and real humans that can answer in a certain way. One [00:49:00] thing I’d throw into that is, um, you know, when Chachi Petit You know, first hit the scene early 2023, late 2022, I had a bit of an existential crisis because, uh, you know, while I have a lot of knowledge and a lot of years working in, uh, audio editing tools, I was asking ChatGPT in the early days, how would you do this?

If I want to compress a voice, what settings would you, how would you do it? And it was coming up with techniques that were better than me. I was like, that’s. Way more efficient. What’s going on here? So, So, um, so there is a chance for AI to, to sort of overtake us and give us just the info we need. Because I think as humans, we, we, we tend to, sometimes we, we beat around the bush a lot and we’re like, well, you need to use this.

And then why don’t you use this tool and then, you know, use the heart limiter as well. And then, you know, Do an overall mastering effect and stuff. And ChatGPT is like, no, just do this, this, and this. And I’m like, actually, you know what? That’s, it’s not a bad idea. Um, I’m, I’m excited about it all. I love the rate of development.

I love the fact that there is a big battle right now between the big companies, Microsoft, [00:50:00] Google, uh, you know, Amazon, and they are throwing out model after model, and there’s a lot of closed source technology. So there’s, GPT 4. 0, Mini presently, uh, the Xanthropics Claude 3. 5 Sonnet, Gemini. Uh, I think they just released some new, uh, smaller models.

Gemma 2, uh, recently came out and there’s the big battle there with open source. So open source, uh, you know, trying to keep things as open as possible. And I love the fact that we’re getting so much technology. In fact, in the last couple of weeks on my AI tools roundups, uh, I’ve covered a lot of Chinese, uh, video makers texting.

to video generators, because they seem to be coming out at a rate of knots. There’s Vidoo. Uh, there’s, everyone’s been talking about Kling as well. Um, and we’re still waiting for OpenAI to say, here’s Sora. And I think now we’ve seen these models, we’ll be like, oh great, but we’ve already, we’ve got Luma’s dream machine and everything else.

Um, so it just keeps things fresh. It keeps things coming. And I think my, you know, final advice on that is just don’t get overwhelmed. Uh, you know, try and find the good tools and, uh, and use them. What you need [00:51:00] to, and make workflows and make templates for your AI as well to, uh, I think like you touched on the start, Jeff, about, you know, if you’re creating a thumbnail or an image or something, give it your voice, give it your style so you can dial things in and it doesn’t just look like a random mid journey generation with stars floating around and yeah.

Jeff Sieh: Nazeem says that he’s using the Journey to Create, uh, Inspiration Voids for architectural designs, which is very, very cool. Thanks, Nazeem. It’s great to see you here. Uh, last question, Mike. This is kind of out of left field, but I was thinking about it when you were talking about all the, your roundup that you do every week.

You run Music Radio Creative, right? That’s a whole business, and you’re doing all this stuff for audio and stuff. And you now have this channel. How do you find time to do all this research, play with all these tools? Uh, do you have like, okay, today’s my AI day. I mean, I’m really interested in your process because, um, your output is phenomenal and you still do all your other stuff.

So, what are your, what does your day look like? How do you do all the stuff that [00:52:00] you do? Because I’m really fascinated about that. Mm

Mike Russell: no magic. I am definitely not superhuman. I don’t like wear a cape or anything like that. The very simple answer to your question, Jeff, is I have an amazing human being next to me as a business partner, as a friend, as a wife, as everything else. Her name is Isabella and it’s all down to her.

Because she’s an amazing business brain behind our YouTube channel. Like you mentioned, our audio business, Music Radio Creative, so she’ll look after a lot of the day to day running of that, the management of a lot of our talent and, and, and bits like that. She really is the glue that keeps things together.

Uh, and then like you say, yes, I think it’s important to have structure, uh, Some days I leave free and open, uh, but some days are extremely structured. Uh, so I’ll have, you know, certain days where it’s non negotiable. I know I’m filming, uh, at this time. And it’s also, I find for me, uh, and I know every creator is different, but if I don’t have like hard marks in the week where something really has to be done, I might not [00:53:00] do it at all.

So things like the weekly roundup right now, this is great to have this in my diary because I, I have to get that out. And, um, And I often find that, um, again, some weeks you’ve really got it, haven’t you? Uh, I’m speaking for myself here. And things work, things flow, and you’re like, yeah, this idea is great.

Oh, I’m going to do this and that and the other. And then other weeks you’re just like, yeah, blank page syndrome. Where, where is everything? And I often find like, when I, when I go away, when I travel, um, not so much to conferences, because sometimes I get inspired from those. But like, if I go on a one week holiday, maybe I can recover for it.

But if you, if you put me in the south of Italy for two weeks with gelato and, and pizza, I’ll be recovering from that for two weeks. So at least I know I come back from Italy, I have to do that weekly roundup on Friday. It’s like, you know, yeah, it’s having the days. And also, you know, I’m, I’m Grateful to have a really great business partner in life.

Jeff Sieh: That’s awesome. That’s awesome. Well, uh, Mike, thank you so much for your time today. This has been amazing. I was so excited to have you on. We’ll have to do it again, uh, soon, but [00:54:00] tell everybody where they can find you, your channels, uh, everything that has to do with Mr. Mike Russell.

Mike Russell: Absolutely. So thanks, Jeff and Connor for having me on. Uh, you can just type my name in. Uh, most places, uh, YouTube especially, Mike Russell, uh, will generally find you. Um, if there’s too much audio when you type my name in and you just want the AI stuff, as we’ve spoken about during this show, just look for Creator Magic.

That’s Creator Magic. Uh, look, look for that. on, on YouTube, um, sign up for the weekly AI tools, uh, email updates, and I’ll, I’ll put the top three AI tools you need this week in, in your inbox.

Jeff Sieh: That’s awesome. Make sure you guys sign up for that. Make sure you subscribe, like and subscribe his channel, because seriously, he’s got some great stuff. Uh, out there. Uh, Gary says, you know, great show this week, guys. Well, thank you so much, Gary. You always are such a great supporter of the show. Connor Brown, where can people find out about the amazing, unsinkable Connor Brown?

Conor Brown: You can find more about me at vacationkingdoms. com and at vacationkingdoms across the social medias. A parting piece [00:55:00] of advice for you, Jeff, when you go to eventually scan yourself for your hologram, make sure you’re wearing pants.

Jeff Sieh: Well, are you sure? Because that wouldn’t be really me. It wouldn’t be me. I’m

Conor Brown: positive.

Jeff Sieh: Okay. Thank you guys so much for watching. We’ll see you guys next week. Bye everybody. Have a great one.

(Visited 91 times, 1 visits today)

Leave A Comment

Your email address will not be published. Required fields are marked *