EP 209: AI as a Creativity Enhancer, not a Creativity Replacement

This is the Everyday AI show, the everyday podcast where we simplify AI and bring its power to your fingertips. Listen daily for practical advice to boost your career, business and everyday life.

Can AI replace our creativity or is it just there to enhance it? Is it just something that's going to get everyone from a 0 to a 5? Or can it bring someone who's a 5 in creativity to a 10? And what does this mean for our future of content consumption? Well, we're going to be talking about those things today and more on Everyday AI. Welcome. My name is Jordan Wilson and I'm your host of Everyday AI. We're a daily live stream podcast and free daily newsletter helping everyday people like you and me not just learn generative AI, but how we can all actually leverage it to grow our companies and to grow our careers. So if you're joining us live, thank you. If you're on the podcast, make sure, as always, check the show notes. So look in the episode description. We have a lot of other great resources for you to check out. So one of those resources is our website every single day. People don't know this. Yeah, there's actually a newsletter written by me, a human. So we recap every single episode. We talk about what's going on in the world of AI news and a whole lot more. It's. I, I'd say it is a free generative AI university. There's 200 plus videos, podcasts, old newsletters.

You can go read on our website.

For free in any category you care about. So you can go, you know, learn about AI in education or AI and entrepreneurship, or whatever you care about. All right, so now let's talk about what's going on in the world of AI news, because there's a lot. All right, here we go. So first, Anthropic is steering away from the voting questions, according to reports. So artificial intelligence startup Anthropic is implementing safeguards for their chatbot Claude to avoid false information, particularly regarding elections. So Anthropic is testing a new technology called Prompt Shield that detects when users of its Gen AI chatbot ask about political topics and then it redirects them to authoritative sources of voting information. So the company's decision to implement this technology was driven by limitations in Claude's ability to provide reliable and up to date real time election information. So we're going to be hearing a lot about that. You know, we've already seen Meta and open AI and some others put in some safeguards as well. All right, speaking of big companies, Gemini 1.0 lasted about a week. Gemini 1.5 is already out. So Google has already updated its newest version of its large language model Gemini. So Google is launching a new and improved version of their large language model Gemini, with an increased context window and improved efficiency for users in the company. So the model is said to outperform OpenAI's GPT4 in most cases and we'll see as benchmarks start to roll out here in the coming weeks. So what's really noteworthy is the impressive context window of 1 million tokens, allowing it to handle larger queries and process more information at once. However, some of those longer context windows seemingly are going to be limited to just businesses, larger businesses and developers, as the standard window will just have a 128k token context window and that's what the general public will use. So that's about 96,000 words of back and forth before Gemini starts to lose its memory. So Gemini is part of the premium Workspace products and has released a limited preview right now to everyone else. So you're not going to see all these full 1.5 capabilities, but they will still be rolling out slowly in the coming days. All right, last but not least, and this is what everyone's talking about. If you've been on the Internet in the last 18 hours, you're seeing Sora. So OpenAI has released some previews of its groundbreaking new tech Sora. So they just released at least some previews for its new text to video model called Sora, which can generate videos up to a minute long based on user input. So OpenAI is sharing its research paper and samples that you can actually go download to gain feedback and is working on ensuring the safety and ethical use of the technology. So the company is collaborating with experts to test for potential harm and will publish safety evaluations. So, so if I have to give you my quick take on this, I did not believe it. If I'm being honest. I use every, I don't know, a single generative AI tool I haven't used. I've used hundreds. Sora. It is mind boggling so far I do not fully understand how it can even produce the level of quality. I actually did a rundown of comparing Runway prompts with sora prompts. Sora, OpenAI shared the prompts that they use to generate these, so I did a side by side comparison. We'll have that in the newsletter as well. But wow. What, what, what a big day. But speaking of creativity, let's talk about that and let's, let's see you know, is, is AI just going to enhance our creativity or could it actually replace us? But don't worry, you don't just have me blabbing on. Today we have an expert guest. So I'm excited to welcome to the show. There we go. Paul. Paul Eater, who is a strategy and data consultant. Paul, thank you so much for joining the Everyday AI show.

Thanks for having me, Jordan. I like to pretend like I know things sometimes, so this is good.

Same. Hey, we're all, we're all just pretending, right? Like there is no generative AI experts out there. We're all learning together. But Paul, maybe, maybe tell us a little bit about, you know, who you are and what you do.

Who am I? Well, I've been in the kind of the data and stats world for the past 20 years. I got a PhD in social psychology which is, I say it's the study of normal people, if that exists. And I have, I basically help organizations become more efficient and effective in a variety of ways. Have been in government consulting for the past 16 years and helped just a variety of government agencies implement new processes and procedures. Some are tech based, some not tech based. So I've, I've dabbled in and out of tech but over the past year or two I've really gotten more into.

The AI space and you know, I'm curious you know, with, with someone with more of a data background. Right. A lot of people, you know, when I came out or you know, quote unquote came out but when it became popularized with text to blank models, you know, a lot of people just went down deeper in their own, you know, kind of niche. Right. So you know, I'm curious even for you, you know, with a data background, you know, kind of what, what pushed you to look at the creative side, you know, as well when generative AI started to popularize.

Well, well this is, this is where I'm kind of a marriage of the left brain and right brain. Right, right. I, I view myself as kind of a data poet and, and so I, it's the, the art meets science aspect of data and I also kind of write poetry on the side and I'm, I kind of do my own journaling and writing and things like that. So what really flipped the switch for me is when the large language models came, came into existence and, and it's also the like the art models that mid journey has and things like that. And it just blew my creative mind and it really made me less of a, you know, AI is a tech world thing to. AI is an Everyday thing. So look at that, look at that plug.

Yeah, no, absolutely. And hey, as a reminder for our live stream audience, thank you for joining us, everyone from Tara and Chrissy, we got people from all over. Daniel joining us from Buffalo. Thank you. But let us know, what questions do you have with AI and creativity and, and let's maybe start at the end. Paul. So, like, what is your overall take on AI? Is it going to replace creatives or is it just going to help people be more creative?

Well, I'll start again, a little bit more of a background. I wrote my dissertation on creativity and how people and organizations can be more creative. So this has been something I've been studying for a while. And over the past year and a half, as this has become more popular, I've gotten in a lot of conversations and probably my most, I guess, interesting conversations on social media, especially LinkedIn, have been in this realm where, you know, there's people strongly on both sides where, hey, these new AI models, these are very creative. They're so creative, I'm afraid they're going to take over the world versus, oh my gosh, my kindergartner could create this in the garage in a couple hours. What's, what's, you know, what's the big deal here? And the conclusion that I've come to is that in a lot of cases, people are confusing the act of creating with creativity itself and they're substituting the concept of speed for creativity. Just because a tool can produce something that is passable or acceptable within 30 seconds, it doesn't mean that something that's passable is creative or that it took creativity to produce it. And, you know, I always go back to what's the academic definition of creativity? And it's really a combination of two things. It's something that's novel, that's also useful. So it has to have both of those characteristics. And where the skeptics come down is that, yes, you know, AI produces useful stuff, but it's not exceptionally novel. And in terms of, you know, some of the words that it produces, it borrows, it uses from other sources, some of the pictures and images it borrows. So in that aspect of it, it's producing something you can use but isn't entirely new.

You know what that concept there, I think is important because, you know, even when we think of different generative AI tools, a lot of them, you know, in order to produce this, they're really just kind of recreating, right? They're recreating what's in their training Data, you know, most companies don't say, but presumably it's anything that's ever been on the open Internet, any image, I mean, any video, you can almost.

And I'll stop you, I think, I think humans do that too, but that's just my bias.

Yeah, yeah. So, but I guess, I guess my thought is, right, so as, as this improves, right? Because anyone that, you know, as an example used mid journey four and then mid journey five and now mid journey six, right. The outputs are obviously getting better and better. So then I guess when it comes to speed and quality, you know, what we just talked about there, what happens right, when not only the speed obviously has caught up, but then what happens when the quality, when it's not, oh, you know, 1 out of 10 are usable, what happens when it's 10 out of 10? Does that change the, I guess, answer to if, you know, creativity is just an enhancer or a replacement?

Well, you know, you're speculating. And so in that regard, I'm still a skeptic, Right. I know AI can produce great things and it's going to produce things that I use all the time, but one of the things that I've been using it for is coding. I've gone from not being a coder to being a Python coder. But one of the frustrations that I have is that it constantly forgets the context or takes out things that I said to include and things like that, just from a usability standpoint. And my personal experience with feels like the bugginess of the output in terms of being anything greater than average isn't quite there yet, nor something I can fathom. So at this point, Even looking at ChatGPT4 versus 3.5, there were researchers that took a look at divergent thinking tests and how well they performed on them. ChatGPT 3.5 performs at about an average person level, whereas ChatGPT force a slightly above average person level. You know, whether will AI get to the super genius level? I don't know. But right now I'm happy with the slightly above average human level and that's a comfortable place for me and. But I don't think it's something where I'm going to take the experts on my consulting teams and replace them with AI because I want people who are experts, not slightly above average people on it.

Sure. And you know, I'm curious, you know, so let's, let's go back to a little bit because you said that you did a, your dissertation on creativity. I want to explore that a Little bit more. So like, like first of all, you know, can you tell us a little bit more about, you know, when this was? But also how do you think, or maybe not. Do you think if you did the same, you know, dissertation today or next year, do you think it would have come out any differently? Do you think that your, you know, like background and research would have changed the outcome of kind of what your dissertation concluded?

Well, you know, my dissertation was in about 2008. Time frame 2007. 2008 time frame 2007, yeah, 2007. I'm going, you're asking to think about timelines here. And it really pertains. A lot of the assessment of creativity was supervisor assessment of employees and how creative those employees were. Now if, if an employee is kind of in the background using some gen AI and they typically weren't the most creative person, maybe their supervisor doesn't know they're using it now and is getting wowed. So maybe there's a slightly difference there, slight difference in how the people would be rated. But, but in the end, what I find is that people who are really not oriented towards being super creative tend to not even care about gen. It's really the creative class who's getting really jazzed about this stuff and using it. So even though it can make people who are generally not creatively oriented get the average, they seem to not really care about it. And that's just my personal experience.

Are you still running in circles trying to figure out how to actually grow your business with AI? Maybe your company has been tinkering with large language models for a year or more, but can't really get traction to find ROI on Genai. Hey, this is Jordan Wilson, host of this very podcast. Companies like Adobe, Microsoft and Nvidia have partnered with us because they trust our expertise in educating the masses around generative AI to get ahead. And some of the most innovative companies in the country hire us to help with their AI strategy and to train hundreds of their employees on how to use Gen AI. So whether you're looking for ChatGPT training for thousands or just need help building your front end AI strategy, you can partner with us too. Just like some of the biggest companies in the world do. Go to your everydayai.com partner to get in contact with our team or you can just click on the partner section of our website. We'll help you stop running in those AI circles and help get your team ahead and build a straight path to ROI on gen AI.

And then so with, I mean, with that in mind, Right. Because these, these, all of these generative AI tools are just getting very, you know, very powerful. Right? Like even, you know we talked a little bit about the new OpenAI Sora model, right that was just released you know about 18 hours ago. You know, do you think that that's going to change this? Right? Do you think that this is going to you know not just like ah, people are, are dabbling and you know someone who's you know not at and they're like ah, not, not too jazzed about it as you said. You know. Do you think that's maybe just because it made people average? But if it does make people or if it has the capabilities to make people above average, do you think then that those non creative people are all of a sudden going to be hey, we're, we're going head first into this, you know, AI creativity thing.

I, I, I don't know whether they'll go head first but their supervisors may think that they can be replaced with it. So if, if you can use Gen AI to get a, an average to slightly above average employee, why would you have a slightly below average employee anymore? And and so I think that's where some people's fear is, is hey, hey, is AI going to replace us? And in the end it could, it could replace you if you're, there are certain leaders who kind of believe all employees are average and replaceable cogs. It's unfortunate but that is, that is really a mindset that some people have and if a leader has that mindset they can make the decision now I can save a lot of money by firing all these people and it not really notice that the complete output has turned to average versus when it wasn't before. So there is going to be some of that. There already has been some of that with some I know notable news organizations trying to generate AI articles and then getting slammed for it because it's such poor quality.

Yeah, let's, let's talk about that. Right? Because I've, I think a lot of people have already started to experience this over the past year just getting bombarded with average written content. Right. Because so many people early on you know jumped on board with chat GPT, you know and unfortunately you know using the free version which is you know, subpar I think especially if you don't know what you're doing. Do you think that as consumers are we going to be hit with an ongoing avalanche of, of just this, this you know, mediocre content and if so what does that even do to the concept of, of creativity. Is, is it going to be too hard for actually creative people to stand out in, in a sea of, of decent.

Well, it. Personality becomes very important that, you know, people will follow. The people that they trust and that they trust aren't robots. So especially if you're in social media, building that audience trust around you as a producer of human content is going to be very, very important. You know, a lot of people, you know, I see this all the time and I'm so doubtful of it. People say, oh, I can tell the difference between AI and human content. I'm like, can you? Can you really? You know, I want to see the definitive tests on that because in the end, you know, if you think that you can spot AI because it's like, oh, it does grammar, right? It must be AI. I, I've had so many, you know, ever since I started talking about AI, so many hundreds of people claiming that all of my articles and posts are AI. I'm like, well, obviously people can't tell because I write my own poorly written articles. Right. You know, and you know, I'm thinking you, you touted the fact that you have human written newsletter. My first, my first skeptical eyebrow raised was is it really human written newsletter? You know, and, and, and just talking about AI makes you think you do everything with the AI. It's such a bias out there now. It's so weird.

Yeah. How do we, you know, because I'm sure a lot of people listening today are maybe creators themselves. You know, they're, they're trying to create content maybe for a small business, for a large business, or maybe to, you know, establish their, their own brand. How can individual creators balance? Right. How can they still leverage generative AI when they need to? But yet what you just said, how can they still have that personality and that trust in that human piece still shine through?

Well, I think the key is storytelling, being able to share personal experiences, things that only you would know and that AI wouldn't know and, and focusing on that aspect of things because then people will trust that it's your own experience. Now can people, you know, tell their experience to chat GPT and have it write it up for them? Well, sure, but, but in the end that is still something that AI is producing that no other person could have AI produce. No one else can have AI produce your own experiences either. So, so there, there's, it's still a differentiator in some ways.

Yeah. And you know, we actually have a ton of great questions, so I want to get to a couple of These. But Mauricio, thank you for this. I think this is a great question. So, Mauricio asking, would you consider effective prompt engineering as a creative tool? Paul, what's your take on that one?

Oh, 100%. You know, my favorite, my favorite guy who really started talking about prompt engineering was Jared Bonilla. He's a AI guy who's been just with ChatGPT and AI from the front and really got me interested in a lot of different things. And he really introduced me to the concept of prompt chaining, which is you have your initial prompt, but what do you do after that when you get your results and maybe think about that from a more strategic perspective as well. So, yeah, absolutely. That's essential for a creative person to not accept what they first get and to continuously tweak and refine and develop their own prompt templates for getting what they need going forward. Absolutely.

Oh, and I, I couldn't agree more. I think anyone that spends a lot of time, hours a day prompting will tell you one of the most important future AI skills, I would say, is communication. Effective communication, clarity in the written word, silly things like typing. Like that's important, right? Like, I think so many younger people aren't great at typing. So guess what? You have to do a lot, you know, when you're working with, with generative AI typing, right?

Well, in some cases I know a lot of people who use like the ChatGPT app and use the voice detect. So typing is less important in that.

In. That's true.

That's true.

Another, Another great question from Andy. Andy, what's up? Thanks for joining. So saying, would you agree that generative AI benefits generalists over specialists by extending the skills and scope of creativity in a wider range of area through upskilling? Great question, Andy. What's your take on that one? Paul?

Oh, well, you know, I definitely, I've actually posted a lot about being kind of a generalist who, who just can't niche down just by my personality. I tend to just kind of jump all over the place. And so for someone like me, it opens up new doors, like becoming a coder that I wouldn't have been able to do otherwise. So I can speak to my personal experience that as a, as a generalist, yeah, it's fantastic. But I know that there are specialists, especially maybe those more in like the, the healthcare space who are really working on AI to help with diagnoses and very specific things like that. And so I can imagine, you know, it's kind of beneficial across the board. You know, generalists just tend to be a little bit more inch deep, mile wide. And now I can be mile deep, mile wide, which is pretty cool.

Yeah, yeah, no it's, yeah, it, it, it definitely changes, I think your outlook on, on not just create like the creative process, but also how you, how you distribute. Right. It, it like what you can create on the front end changes everything that your company can or maybe shouldn't do on, on the back end. Our audience is on fire today. So many great questions. It's a Friday, right?

Like normally fire starters just say, you know, so, so I'm setting fires everywhere.

I love it. All right, well here's, here's a tough one. And Tara, thank you for your question. So Tara said, how can I balance the enforcement of watermark attributions with the principles of fair use, ensuring that content remains accessible for educational review and commentary purposes without infringing on creators? Right. This is like a trillion dollar question right now. What's your take on this one, Paul?

This is one that I've written about, you know, quite a few times on LinkedIn and get a lot of people who are very strong in the, the copyright protection area. But my take on this is a little bit more, I guess we'll say libertarian, freedom focused of if, if you're using something that's copying someone else's work and they can prove it, they have the legal system to help them do that. Some something is not copied and, and not, and not unattributed unless the lawyers say so and the judge says so. And, and so it's, it's kind of, I'm more of a, you know, a legal system absolutist on this, that if you think you're being infringed, take it to court.

Yeah, yeah. And that will be, I think, a topic that at least here in the everyday AI show we'll be talking about non stop. I've talked, you know, probably way too much about how 2024 we're going to see so many court cases for these exact things. Copyright infringement, IP watermarks, you know, it's, it's gonna be messy, but it's gonna be messy.

But OpenAI can afford really expensive lawyers, so they're gonna win.

Yeah, I said, I, I said they probably have like the 92 dream team of lawyers.

Yeah, yeah. There's no way they don't win this.

Yeah, yeah, yeah. Or just settle everything. Right?

Yeah.

So Paul, another good question here from Daniel. And especially I think with your background, you know, your dissertation on creativity, I would love to Hear your answer on this. So Daniel asking, do you think generative AI will be its own course in school or will it be become part of curriculums in every class?

I have advocated that just, just from the start, like teachers immediately all over the place started fretting. Now I don't know if people are plagiarizing. I don't know, you know, you know, you know how to grade these kids papers. And you know, my, even my own son had something sent back to him, saying this sounds too much like AI, rewrite it, you know, and in the end, you know, the education system can't place itself in a place where it's an opposition to the Zeitgeist. I'm the guy who is the key word that I get bonus points for. It has to embrace it. And if it really wants to train children for the future, incorporate it into classes as much as possible. And you know, teachers can easily develop rubrics where they look at, hey, what was my initial prompt? How did I develop a prompt chain based on that? How did I revise the final product? You can ask for all stages of the process. If you're worried about the final product, product is going to be plagiarized. Ask them to show you each phase. And that's, I mean, that's in my mind what's going to have to happen if we want our kids to know this stuff and succeed with it. Love that.

Great, great question, Daniel. Great answer, Paul. And hey, FYI, teachers, admin everywhere out there. I don't say this enough. There is no such thing as AI content detection. It literally doesn't exist. I've probably spent 30 hours, you know, trying out everyone. Not, not a thing. Y' all just putting that out there. So, Paul, we've, we've talked about a lot. We've talked about the active of creativity versus, you know, the act of creating something versus creativity. We've talked a little bit about even your own personal experience, you know, with, with coding and with your background, you know, your dissertation in creativity. Right. So as, as we kind of wrap up here, you know, because I'm sure there's a lot of people who are constantly thinking about how AI does impact them, either as creators or potential creators. So maybe what's your one takeaway, piece, you know, that you can give to people that, you know, that they can if they're interested in becoming more creative with generative AI, how do they do it and what should they keep in mind?

I use AI as a starting point, but not a finishing point. If you're if what you're doing is taking AI content and cutting and pasting it and giving it to someone, you're going to be average. If what you're doing is taking that, what you initially get, adding your flavor, enhancing it, you know, even if you're doing AI art or AI video, incorporating that with something else that you're doing in, in a more nuanced way, that is going to be just more beneficial for society and if you're in social media for your audience than just any kind of cutting and pasting any day. So that's, that's, that's what I advocate.

Great, great advice. I love that. You know, it's not a starting point. It's a finishing point.

Well, that's, that's our finishing the other way.

Oh, yeah, sorry.

Yeah.

Thank you. Thank you for correcting me. Hey, it's Friday for all of us, right? Like, hey, that just shows you I'm not an A.I. jordan. I'm a human. I make mistakes. Right? But, hey, we covered, covered so much here. Paul, thank you so much for joining the Everyday AI Show. We really appreciate it.

All right, thank you very much for having me. Appreciate it.

And hey, as a reminder that we covered a lot, so make sure you go to your Everyday AI dot com. Sign up for that free daily newsletter. We're going to be recapping what Paul and I talked about today as well as a lot of other things, all that news we talked about, and a lot more. Thank you for tuning in and we'll see you back for another episode of Everyday AI. Thanks, y'. All.

And that's a wrap for today's edition of Everyday AI. Thanks for joining us. If you enjoyed this episode, please subscribe and leave us a rating. It helps keep us going for a little more AI magic. Visit your everydayai.com and sign up to our daily newsletter so you don't get left behind. Go break some barriers and we'll see you next time.