EP 180: AI News That Matters - Jan. 8th, 2024

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.

Keeping up with everything in Generative AI is like a job and a half. I mean, trust us, this is literally what we do at Everyday AI and there's always so much news, tools, events, happenings, acquisitions, it's, it's even hard for us to keep up. So we're doing something a little bit new@everyday AI. This is one of the first new things that we're talking about here in 2024, where we're going to take each Monday and talk about AI news that actually matters. And we're going to get away from the bullet points and, you know, here's 50 news stories. We're going to talk about a handful of stories in depth and really hopefully help you talk about what they mean to help you grow your company and to grow your career. Hey, what's going on? If you're new here, good to meet you. My name is Jordan Wilson and I am the host of Everyday AI. Everyday is a daily live stream podcast and free daily newsletter helping everyday people like you and me learn and leverage Generative AI to grow our companies and to grow our careers. So if you are joining on the podcast, appreciate that, would love if you, if you could subscribe and leave us a rating and tell your friends about it. And for those of you joining on the live stream, thank you as always, like Brian joining from Minnesota, Mike Forgey and Woozy Rogers, thank you all for joining us. But if you are joining us on the podcast, we always leave other links that you can come back, join the live stream, ask questions and connect with a thriving community of AI enthusiasts. All right, so without further ado. Well, no, actually I do have to tell you, if you haven't already, go to your everydayai.com there's some new stuff there you probably haven't seen, but we're going to be announcing a lot of new updates this week. But you can sign up for the daily newsletter which you should be doing@your everydayai.com but also if you didn't know, we have now more than 170 episodes that you can go listen to. You know, I get messages and emails all the time saying, hey, Jordan, how can I learn about generative AI for sales? Or hey, how can I learn, you know, about the ethics of generative AI? And you can just go on our website and we have these different AI learning tracks, probably more than 30 of them. So with more than 170 episodes, we've covered literally just about everything. So make sure you go to that. And you can also, on our website, read every single email newsletter. So once you subscribe, you can, you can go back and read so much. So I cannot emphasize enough, even with Keeping up with AI News and what Matters. Those are two fantastic resources looking at our backlog of more than 170 episodes and going back and reading every single newsletter as well. So with that, let's talk about what's not just going on in the world of AI news, but what, why it matters. Right, so couple. Couple news stories that we wanted to look at a little more in depth. All right. And yes, we did cover these in our newsletter. We covered the daily news, but I wanted to dive in a little deeper. And also, if you're joining us live, like Nina from Denmark or like Shannon, please let me know, let me know what you think of these different topics that we're talking about and what questions you have. I'll try to answer some of them live if I can. All right, so the first one, an article here from IEEE Spectrum, is talking about how generative AI, specifically Mid Journey, has a plagiarism problem. Yeah, a visual one. So we're going to talk about this here in depth and show a couple of examples. But also tomorrow, this is really what we're going to be talking about in line with the New York times in the OpenAI ongoing lawsuit, but in this case, we're talking about Mid Journey. So we cover this in the newsletter. But Mid Journey finally released in Alpha, but they're V6. Right. So mid Journey actually puts out a lot of great, fantastic updates. So we went from, you know, you go from V4 to V5.5.152. Right. So mid Journey pushes out every, you know, month or two, these incremental updates, but they're actually very impactful. And one thing that a lot of people have been talking about, especially since this new V6 release, is that, hey, a lot of these image generations are looking exactly like copyrighted images, specifically movie stills. All right, so have to shout out the, the people that have put, put the work in. So we'll, we'll link to them in the newsletter. There's been a couple of people that have really been going in depth, looking at different movies, running different outputs inside Mid Journey. We, we replicated some ourselves, but we're not going to share them. And more on that here in a second. So let's take a look. This is an example here. So if you are listening on the podcast, go ahead, check your show notes. Come look at this on the screen. But. So this prompt says, famous movie screen caps superhero, and then some technical things about the. The aspect ratio. But just by putting in the prompt, famous movie screen caps superheroes, I am getting back images from different superhero movies, right? And that begs the question, what should AI image generators be doing? When I put in a very vague prompt like that, Famous movie screen caps superheroes. And I'm getting the results. I'm getting back. I have a very lifelike image of. Of Thor, right, From the Avengers and then Superman, and then another one that has different images of Batman and the Flash. These are all obviously copywritten, right? It's something. It's. It's kind of like the largest elephant in the room that no one wants to talk about, especially when it comes to AI Arts generators, specifically in Mid Journey, right? These are copywritten images. H. It's troublesome, right? And, yeah, we've been using Mid Journey essentially since it came out. And we have noticed specifically with V6, not just. It is almost impossible, y'. All, it is almost impossible to tell the difference now between these AI generated images and real life. And that makes it especially troublesome when you get generations like this when you're not even asking, right? In this prompt, the prompt is not asking for images of, you know, Thor or images of Superman or images of Batman. The prompt is only asking for famous movie screen caps superhero, right? Kind of wild. And then we're getting back images that are clearly from, you know, big studios that have the ownership rights to these, right? And then you have people publishing these works all over the Internet and also using Mid Journey and other AI image generators for commercial product projects. So the waters are going to get very muddy, right? I've. I've got some hot takes coming in tomorrow's show regarding the New York Times and the open AI situation. Very similar. So the New York Times is suing open AI and technically Microsoft for similar copyright infringement. But I'm actually surprised that this Mid journey, especially with V6, y', all, because this new V6 update, the realism is through the roof, right? I used to kind of be a photographer in a form of life. I've taken at least 250 to 300,000 images with a DSLR camera. So you could say I have a decent eye for photography or graphic design or something like that. I cannot. With. With this new v6 update, it is nearly impossible even for a trained former photographer like myself, to tell the difference between what is Real and what is AI generated. And that becomes especially problematic right when we start to see AI image generators like this create just so highly replicated almost pixel for pixel, like duplications from movies. All right, so here I've got, got another something to. To look at here and something to probably you might want to re look at your use of Mid Journey, FYI. So yeah, there's, there's been a lot of talk on the Internet, you know, Reddit, Twitter, different, different news articles like the one that I shared and that we'll put in the newsletter. But you might want to go look at Mid Journey's new Terms of Service. A little troubling. Yeah, a little, little troubling. Ready? So the updated Mid Journey terms of service say this. You may not use the service to try to violate the intellectual property rights of others, including copyright, patent or trademark rights. Doing so may subject you to penalties, including legal action or a permanent ban from the service. We reserve the right to investigate complaints or reported violations of our Terms of Service and to take any action we deem appropriate, including but not limited to reporting any suspected unlawful activity to law enforcement officials, regulators or other third parties. Y', all, I don't know if, if you understand what's going on here, right? But we've been using Mid Journey, like I said, essentially since it came out. And I don't know if our team at Everyday AI can continue to use Mid Journey with these terms, right? And here's exactly why it says you may not use the service to try to violate the intellectual property rights of others. That right there, in my opinion goes against really everything that a lot of AI image generators are trained on, right? AI image generators are trained on copywritten materials, right? And even in this example prompt. So again, if you're listening on the podcast, you've got to see this. So just click click your show notes, go join the live or you know, look at the website. But even when you are not trying to violate terms, right, like this prompt that just says famous movie screen caps superheroes, there is no intent in that prompt to violate any copy copyright images, right? There's not yet the return you get, as you can see on the screen if you're joining us live, is copyrighted materials.

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 gen AI. 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.

I'm curious, for those of you on the live show that are using Mid Journey, did you know about this new updated Terms of Service? I didn't see exactly when these terms were updated. Presumably it was around this V6 release. I'll have to check and go into the the archive.org and start looking at different screenshots of their Terms of Service to see exactly when this language was updated. But essentially what Mid Journey is doing here is passing the buck on to users, right? Which I understand from the company's perspective. However, I don't, I, I don't really see how any now Mid Journey user can use Mid Journey without violating intellectual property rights. Right. Because as you saw on this example, the prompt is not even trying to do that. It's just asking for photos or screen caps of superheroes and you get owned images from others. Not a good look. Not a good look. Yeah. Gert here saying, I got an almost perfect Superman from Dolly today after it explicitly said this one is unique and it doesn't resemble a real superhero. Yeah, yeah. So it's not just Mid Journey, you know, all different AI image generators, platform, AI image generating platforms are running into these issues, right? I haven't used Adobe's products as much, but Adobe is really pushing, I guess, images that it has the rights to when it trains its AI image model. So we'll see. But you know, when we look ahead in 2024, we haven't really seen this marquee lawsuit yet when it comes to AI image generators. Right? Like we have the marquee lawsuit now with New York Times, you know, one of the biggest names in the world in publishing versus OpenAI. So you have the two giants there going head to head. We haven't really seen that marquee kind of AI image generating lawsuit. Yet, But I do feel we're going to see it soon, especially After Mid Journeys v6 updates. These new terms of service, it's something to keep an eye on. All right, let's talk about the next piece of AI news that I think it's worth keeping your eye on. AI powered appliances are coming. All right, so Samsung is actually planning this week to unveil a new fridge. And you're probably like, all right, Jordan, what are we talking about fridges for on the everyday AI show? Isn't this supposed to be about generative AI? Well, they are releasing a bespoke four door flex refrigerator with an AI family hub. What that means is this has an internal camera that can identify up to 33 different food items and suggests recipes based on what you have in your fridge. All right, here's why I think this is news that matters. This isn't the first AI powered device, you know, hardware device that's available. It's not. There's so many, and we're going to talk about those some more of those here in a second. But this fridge from Samsung, as small or as insignificant as this may sound, I think this might be one of those first big steps into bringing AI into your home without even necessarily choosing it or knowing about it. Right. Even if you're just a tech geek or maybe you just like, you know, great appliances, if you see this Samsung fridge, you might just say, okay, yeah, I like this. I'm going to buy this fridge, right? We've needed a new fridge. I want to splurge on something nice. I'm going to buy this new fridge. I think 2024 is going to be one of those years where it starts to, we start to bring generative AI into our lives in ways that we didn't have in 2023 and maybe in ways that we're not even realizing. Right. So I think this might be the signal of one of those first big steps of the average consumer bringing generative AI into their life and not even knowing it. Right? I mean, yes, if you're using smart assistants like Alexa or Siri or something like that, yeah, you've been using AI technically for a while. But I think it's going to be important to look at what happens with these AI smart devices, smart appliances that are using generative AI. So, yeah, speaking of that, that brings us to our next big AI story to keep an eye on. So big AI announcements are definitely going to be coming to CES this week. So CES is the consumer electronic show and it builds itself as the most powerful tech event in the world. All right, so if you haven't heard of ces, I mean, I follow it every year, but I'm a dork. Any other dorks out there that follow CES like the Super Bowl? It's just me. All right, but it is in its 58th year. So CES kicks off tomorrow in Las Vegas and little preview here, shown on the screen from CNN Business, of what's to come. Spoiler alert. It's a lot of AI. A lot of AI announcements are expected at ces, which, like I said, starts tomorrow, kicks off in Las Vegas and runs through Friday. So a couple of things that have already kind of been leaked or previewed of, things that we're going to see, some new AI announcements coming from ces. So companion robots with chat GPT enabled, right? Which could be a fun thing. It could be a very bad thing, right? This is another thing if, I mean, if you're a parent and you see, you know, oh, chat GPT enabled robot for kids, do you want it? I don't know, right? Like, not just accidentally jailbreaking GPT, but I think so many, you know, whether it's software companies or hardware companies, and now you, you know, have the easy ability to kind of create your own, you know, custom GPTs and embed them on your website. I don't think all these other companies that are using generative AI, especially, you know, tapping into OpenAI's API for the GPT technology, I don't think everyone understands fully what GPT is capable of. Right. Or what safeguards that different hardware manufacturers or, you know, third party iOS app developers that are tapping into OpenAI. I don't know if all of these parties fully understand what GPT is capable of. As an example, if I had a child, I probably wouldn't want them to have a GPT powered companion robot, because you can get GPT and other large language models to go off track, right? You can jailbreak them, you can get them to say and act in very bad ways that they were not intended to. So I think that's something that we should probably be keeping an eye on when we're talking about news that matters. All right, so some other things happening at ces, smart belts for the visually impaired. All right? And also kind of like what we just talked about, AI driven home appliances. Also, what's expected at CES is the marriage of more voice technology and generative AI systems, which I am here for that I've just really got gone hard on different, you know, Platforms like Man, Alexa and Siri for being smart assistants are not very smart. So I am very much looking forward to when you get these kind of voice, voice technologies that are powered by generative AI and not necessarily the systems that have been powering them since we've been using them for now, you know, nearly a decade, you know, eight or nine years. So I think that's a good thing. Yeah, Mike here with a comment. Smart appliances are stupid says, I hate my smart tv. It was obsolete in a couple of years. Yeah, but that's, I mean, that's, that's where we see now, Mike and, and everyone else listening. I think 2024 is going to be the year where, you know, your traditional quote unquote, smart appliances, smart hardware, Smart software are going to stop trying to create this technology in house, which a lot of companies have been trying to do for dec, and they're going to rely on partnering with different generative AI models, different large language models. Right. So presumably we're going to be seeing a lot more big partnerships with OpenAI and using their GPT model, we're probably going to be seeing a lot more partnerships with Google and they're, you know, using their Gemini Pro or when Gemini Ultra model comes out in Bard in 2024, I think you're going to be seeing a lot more partnerships with them. And as well as Anthropic. Anthropic's clogged too. So I think that this is going to be the beginning, I think, where smart assistants, you know, that we're bringing into our homes, like these different devices that we're probably going to see announced at ces, that we're actually going to say, oh, wow, this is actually smart. You know, I don't know about all of you all, but when I first, our team first started using the GPT technology in late 2020, right, when it was first announced or it was first commercially available in a lot of popular products like, you know, copy AI Jarvis now called Jasper. So late 2020, early 2021, that's when our team and you know, myself personally, started using, you know, these generative AI technologies and was like, wow, this is AI. That's actually really good. Right? This was almost now three years ago saying this technology puts Siri and Alexa to shame. So I think keep an eye on what comes out at CES this week, you know, again starting tomorrow. But I think as these companies start to leverage different large language models, we're finally going to get smart appliances that are not stupid, I think. So we'll see. All right, and our last big piece of AI news that matters for this week, Monday, January 8, is Deloitte going all in on generative AI. So, sharing the story from the Financial Times here and we'll have more about it in our daily newsletter as well. But Deloitte has officially launched its new AI chatbot named Pair D and they've launched it internally to to 75, 000 employees in Europe and the Middle east, as well as the Europe and the Middle East. So this chatbot is designed to help Deloitte employees or to assist them in creating everything that they do so from PowerPoint presentations, writing emails, coding, productivity tasks, etc. Right. So here's an interesting note though in this announcement, but staff are still advised and encouraged to verify the chatbots out the chat bots output for accuracy, meaning that the AI is obviously still, still evolving and not always entirely reliable. That piece is very important and it's, it's kind of that, oh, hey, make sure you, you know, cross your T's and dot your eyes. But I don't think most employees, and I'm not just talking about Deloitte and their new Pair D large language model, but I'd say most employees that are using large language models. So whether we're talking about Microsoft Copilot or maybe you have OpenAI their enterprise chatgpt, or maybe you're using Bard Claude, I don't know. But I'd say most companies that are leveraging large language models are not properly checking their work. They're not. Here's why. A lot of times, depending on what large language model you're using, it can almost be as time consuming to double check the output quality of a large language model than it is to do it yourself. Right. Obviously, if the large language model provides citations, which a lot of them are now, especially perplexity. So hey, shout out perplexity. They've been ahead of the curve and they've been doing that since day one. But a lot of times it takes just as long to go and double check a large language model's work if it doesn't give you proper citations. So I don't know if that's going to happen or if it's going to continue to happen, especially as now large, you know, house, house household companies, name brand companies are starting to use large language models now out in public. All right, hey, look at this. Josh, Josh, Josh, comment here. Saying speaking of Deloitte's ey, Ernst and Young publicly announced in September that they're investing $1.4 billion in AI. Yeah, Josh, thank you. That's a great, A great point there. And kind of what I wanted to get to next is that in the management consultant industry, there's has been this phase of. All right, layoffs, right? Mass layoffs. Some big management consultants, consulting companies had tens of thousands of layoffs in 2023. Some not so much. I think Deloitte only at least announced around 1200 layoffs in 2023, which is not a lot. But Deloitte is saying that they are using AI to avoid layoffs, which I think is a smart move for your Fortune 100, Fortune 500 companies. When you are putting that big kind of hat in the ring, so to speak, your public hat in the ring saying we are here to use AI, I think it is wise to say we are doing this to avoid layoffs and not necessarily when certain companies say, hey, we are laying off or we're not hiring for these 8,000 jobs because of AI. A big difference there. But I think this is going to be a trend in the management consultant industry. And I've been very hard on the different management consultant companies, you know, not just the big four, but all of them, because I think they've been dragging down their feet on this. Right. So I, I had a show a couple weeks ago, so I was a featured speaker out at the AI Summit in New York City, one of the largest AI summits in the world. And kind of in between the time when I was, you know, not speaking or preparing for my speech on how to make chat GPT sound more human, which was a fun, fun speech to put together, by the way. But I was speaking to a lot of different management consultant companies because a lot of them were obviously represented by, at this AI Summit. You know, they had big exhibits. And I was actually shocked at the amount of, you know, outside of the big four, but your next tier of management consulting companies, I was shocked at even the household names, names that I knew. You know, just kind of. We'll. We'll say kind of tier two management consultant companies. How many of them did not even have an AI plan yet? You know, not everyone is like Deloitte or ey, like what Josh is saying here, investing billions of dollars into generative AI. But you also have management consultant companies that are kind of just throwing money at it and saying, yeah, we're doing something. We're doing something about this, probably to appease their larger. Their larger clients that are investing millions of dollars annually with their management consultant companies to say, hey, you shouldn't Abandon us. You shouldn't go use, you know, or build your own large language model to replace us. We're investing money and resources in generative AI. Right. I like what Accenture is doing. They were one of the first to appoint a chief AI officer, which was smart. I like what Deloitte is doing. Deloitte has been big on AI. You know, ey like what Josh said, EY was actually one of the presenting sponsors at this AI summit. And I got to get some behind the scenes on what they're working on with, with their generative AI and just, you know, deep learning, machine learning efforts. But I think you just have a couple leaders and then you have the rest of the pack. And the rest of the pack I think is falling behind. So if you are a business owner out there or if you work at a big company who uses a management consultant company, I think it's extremely important to not just ask them what they are doing in terms of leveraging generative AI and large language models for your company, but also, what does it mean? What does it mean? Are you going to be spending less time? Are you going to be spending more time? How are you going to be verifying outputs that this large language model gives you? Right? So similarly to how this news article from Financial Times mentions that Deloitte is still encouraging its staff to verify this pair d kind of chatbots outputs. But if you're a business owner, a leader in a, in a division and you're working with a management consultant company, you need to grill them. You need to grill them on how they are using generative AI, on how they're using large language models. Because here's the thing, traditionally when, you know, large language models kind of first burst onto the scene with ChatGPT, that was again, they've been out for years before that, but that's when they were kind of popularized, right? But essentially what you had smart people started saying, all right, well what's the need now for management consultants? You know, if this large language model can, you know, correctly computer tens of thousands of rows of data in a spreadsheet and it can look at all this different, you know, case law and how, you know, different management consultant companies have, have taken on big, you know, restructuring projects over the past 10 years. Some people have said, all right, well what's the need for large or for management consultant companies? Right? So if you are working with one, you should be asking them questions. I'm not telling you to dump them. That's not what I'm saying. But you need to understand. Right. Because you are going to see. You know, I think Deloitte, Deloitte EY and Accenture are probably three of the first ones to really have a firm public stance on their use of AI. But you need to understand how they're using it. You need to ask them questions and you need to understand ultimately what outputs are being used verbatim from a large language model. What sources are they tapping into? Are they using OpenAI? Do they have anything proprietary between which, gosh, I hope, yes, most of them probably do. But you know, what proprietary methods or internal training are they doing? Internal fine tuning? What are they doing on these large language models before they're taking these outputs and advising your company? It's extremely important to keep an eye on or like what Shannon says. This is where PPP comes in. Yeah, take our free Prime Prompt Polish course. It's coming back tomorrow. We took some time off from the holidays, but our free PPP course is coming back tomorrow. So holler at me if you want access to that. All right, that is a wrap, y', all, for our first, our first rendition of this new weekly feature of AI News that Matters. If you're listening on the podcast, please, I cannot stress this enough because you know, we have nice, easy access to our live stream audience here, as always. And I can't, I can't tell you live stream audience here on LinkedIn and I guess some people on YouTube how much we appreciate you tuning in. But hey, if you're listening on the podcast, let me know when we start doing this this Monday. AI News that Matters recap. Please let me know what is going to be helpful for you. We make the show for you, so please give me your feedback. This is our first, first episode of this on a Monday. But here's, here's what happens. I feel sometimes by Friday, so much has happened in AI News, it feels like you are hit with, with, with a ton of bricks, right? Especially if you are someone within your organization who is responsible, you know, for, for making sure that your company is, is keeping up with generative AI. You know, I've gotten so many emails and messages saying, hey, Jordan, from, from listening to everyday AI and, you know, being a part of this community, I'm now in charge of, you know, AI implementation in my company. So let me know how we can better help you not just talk about what's happening with this AI news, but why it matters. Right? So that's what this, this Monday segment is supposed to be, is to take a look beyond the headlines, right? Beyond the spend, beyond the pr. And to say, here's what it actually means, here's what these big news stories, here's how they're going to impact us all, here's how they're going to follow us home in our fridges, right? Here's how. You know, if you're using AI image generators like Mid Journey, you need to be aware of these things, right? If you're working with a management consultant company that's now announcing, you know, these different gen AI and large language model, you know, implementations or you know, their own products, here's what you should know, here's what you should be asking them. So I hope this was helpful. And please, speaking of, you know, we just talk about Mid Journey and some, some legal things that you should be aware of. But also, hey, tomorrow, make sure to tune in. We're going to be talking about the New York Times versus OpenAI and the huge AI implications that no one is talking about. No one is talking about this, right? So hey, I was a former journalist for about eight years, obviously have an AI media company now. So I have some interesting insights into this whole New York Times vs OpenAI lawsuit, what it's going to mean and some big implications that no one is talking about. I hope today's episode was helpful. If so, please go to your everydayai.com Sign up for that free daily newsletter. Reply to it, ask me a question. We get to everything. So thanks for joining us. Hope to see you back tomorrow and every day for more 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.