EP 285: AI News That Matters - June 3rd 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.

Yes, there's even more. Nvidia chips, AMD chips. Everyone's joining forces against Nvidia, Microsoft, Meta, everyone. There is so much happening in AI news this week. It is hard to keep up with. But you're in the right place. We're going to do that today on Everyday AI. What's going on, y'? All? My name is Jordan Wilson and I'm the host of Everyday AI. And this is your place. This is your place to stay up to date with generative AI news that helps you grow your company and grow your career. So if you're new here, thank you for joining us. Almost every single Monday, we bring you the AI news that matters. That means, you know, we sift through hundreds, literally hundreds of AI news stories each and every week and bring you the handful or so that you need to know about in order to grow your company and to grow your career. So if you are brand new here, thank you so much for joining us. We do this for you. A daily live stream, podcast and free daily newsletter to help us all learn and leverage generative AI. So with that, let's just jump straight into it and. Well, first let me give you a reminder because if you haven't already, make sure to go to your everyday AI.com and sign up for the free daily newsletter. Because, yeah, we're going to be bullet pointing a lot of AI happenings, but this is not everything. So you need to be reading our newsletter each and every day. It's free. So make sure to go to your everyday AI.com and sign up for that free daily newsletter. All right, so let's just jump straight into it and bring you the AI news that matters for the week of June 3rd. All right. Hey, if you are joining us live like Rolando here and Tara and Brian, first of all, thank you as always for showing up, but I'd love to hear your takes on the AI news that matters this week or if there's something maybe we missed. What are you keeping an eye out? What are you following or what do you want to know more about? All right, so let's go ahead and start off with the big news. Yes, Nvidia has unveiled its next generation AI chip architecture called Ruben. So Nvidia has unveiled its new AI chip architecture named Rubin, following the previous model, which is Blackwell. It's weird calling Blackwell a previous model because this has been a rapid development between Blackwell to Rubin within less than three, three months. This kind of showcases Nvidia's accelerated pace in AI chip advancement. So this just was announced at Computex and Nvidia is committing to releasing new AI chip models annually. A shift from its previous two year update timeline. So this new AI chip called Ruben, will introduce new GPUs and a central processor called Vera to enhance AI system training and deployment. Competitors like AMD and Intel are striving to keep up with Nvidia in the AI chip market, while companies such as Microsoft, Google and Amazon are also vying for the top spot. The unveiling of Rubin signifies a significant shift in computing driving the next wave of technological advancements, according to Nvidia CEO Jensen Wong. Also, Nvidia announced G Assist, a live AI assistant to assist video gamers that runs on Nvidia's latest hardware. Am I the only person in their mid-30s that doesn't play video games? I feel it's just me, but pretty, pretty cool advancements here from Nvidia. And this is, I mean if this doesn't show you the pace of generative AI, I don't know what, what, what will or what does, right? So I was at Nvidia's GTC conference in San Jose just about three months ago where they announced their new Blackwell chip. And here we are already three months into this and they are already unveiling its successor. Even though so many people haven't even gotten their hands on this new Blackwell chip. So if you are new to kind of GPUs or the AI chips and you're wondering why the heck does this matter? Well, all of the technology, all of the generative AI technology that we essentially use uses these GPU chips. So you know, think about, you know, a chat GPT or a mid journey or a Runway or a copilot, right? So if you're using Windows Copilot, all of these generative AI technologies need these GPU chips and the faster in more capable that these chips get, the more powerful our generative AI gets on the back end. So we've seen literally mind boggling, a mind boggling pace of updates, you know. So here we said normally, you know, Nvidia's on this two year cycle of releasing their next gpu. And now they announced it is going to be annually. And here we are just three months after Blackwell, which blew everyone's minds. Here we are with Reuben. So geez, it's, it's, it's hard, it's hard to keep up. I don't know if anyone else you know is, is left scratching their heads or just me. All right, so speaking of Nvidia. Well, our next piece of AI news is essentially everyone except Nvidia is joining together and banning together to challenge Nvidia's dominance in chip interconnect technology. All right, let's break down what that means. So in response to Nvidia's strong position in high speed chip interconnect technology, major players like amd, Intel, Google, Google Meta, Microsoft, like everyone. So they've all come together and have formed this group called the Ultra Accelerator Link or UA Link Promoter Group. All right, so the primary goal of UA Link is to establish an open industry standard for high speed interconnections between AI processors, providing a competitive alternative to Nvidia's NV link technology. Yeah, so essentially NV Link is really the only major player out there. So this is one of the reasons that Nvidia kind of has a, a just grip on the entire industry. So here you are, you know, I mean, when was the last time you saw, you know, Google, Microsoft, Meta, amd, Intel, all these groups come together and essentially saying, hey, we need to form an alliance here. Right. So this, this, you know, if you're a sports fan kind of like me, I think of when LeBron James and Dwyane Wade and Chris Bosh kind of formed together this super team. So I mean here's what you have there is with this new UA Link, Nvidia is so dominant and so far ahead of everyone else. Not just with its GPU technology, but with its NV link. Now you have this group of unlike. Right, Like a very unlikely combining of all of these different people. Yeah, I mean just even look at Google and Microsoft. When was the last time Google and Microsoft came together on something, let alone amd, Intel, Meta, all of these. So anyways, the group plans to develop a specification for a high speed, low latency interconnect to facilitate communication between accelerators and switches in AI computing pods. So again, aiming to challenge Nvidia's market dominance. So this 1.0 specification is expected to be released in the third quarter of this year or with potential product implementation by 2026, making a significant step in the collaborative efforts to rival Nvidia's technology. So through collaboration with the Ultra Ethernet Consortium, which is the uec, UA Link seeks to address challenges posed by Nvidia's technology, emphasizing the importance of improving efficiencies and performance in pod cluster architectures. Yeah, I, I Pretty much understand this, but it is kind of hard to fathom. But, you know, if you aren't a super technical person, right? If, if you are an everyday person, essentially you have these racks of GPUs, right? And they all kind of have to work together. So Nvidia's technology with this NV link is kind of leap years, or has been leap years ahead of everyone else. So it's, it's kind of forced a lot of big companies, you know, so if you're trying to run your own generative AI, if you're trying to run your own models, you don't always have choices but to use. I mean, Nvidia's GPU chips are more powerful than everyone else's, and then you kind of got to use their NV link. So pretty strategic move here from everyone else except Nvidia, you know, and this kind of came out about Thursday. And then you had Nvidia just come in and just slap everyone, you know, yesterday with this new announcement at Computex saying, all right, well, hey, you guys are all going to form together. You know how our Blackwell chip is better than every other chip. Here's the chip that's going to replace Blackwell, which is already leaps and bounds ahead of everyone. So pretty interesting developments here going on in the chip space. Yeah. So, hey, we. We got a live. A live crew here from the live stream. So thanks, everyone. We got Josh joining and, and Liz and Jennifer. Thanks y' all for joining. But what do y' all think of this chip race? Is it. Is it interesting? Is it exciting? Personally, even though I'm not a super technical person, I mean, you have to pay attention to this because this is the future of, you know, compute, Right? So when we talk about, you know, if your company is running its own models, if your company is making decisions on what generative AI provider to choose, I mean, you have to at least be aware of what's powering it behind the scenes. So it has been traditionally Nvidia, but now, you know, essentially everyone else came together and said, hey, we can't compete. So we have to come up with unique ways to kind of band our resources together so we can compete with Nvidia. So pretty interesting. All right, speaking of pretty interesting here. Well, yeah, even. Even more. Even more AI. So AI chips. So AMD also has announced new AI chips to try to compete with Nvidia. So AMD has announced new AI chips for laptops and desktops at the Computex Tech Conference, which is going on this week, with plans for data center chips and server processors in future. So AMD is focusing on AI and aims to establish itself as a leader in the industry, which is wild to think that because AMD has always been a top name. But you know, Nvidia has just separated itself so far from everyone else that you have these traditional companies now, you know, really just trying to release the, the latest and greatest chip to, to be relevant, which is really hard. So they've announced the new Ryzen AI 300 series for laptops in the Ryzen 9000 series for desktops, expected to launch in July. So AMD also announced plans to release new AI chips every year to compete with Nvidia. All right, let's get away from chips. Let's talk about text to sound. All right, so popular AI text to speech company eleven Labs is now releasing text to sound. So yeah, they've been one of the leaders in text to, to voice technology. 11 Labs has for the last probably two years. So now they have released their newest update to 11Labs that brings text to sound. So 11Labs has introduced a new AI audio model that can generate sound effects, instrumental tracks, soundscapes and character voices just from a text prompt. So the model is designed to cater to content creators, video game developers and film and TV studios, providing a convenient way to create more immersive content. So the AI model does not have content filtration, allowing the generation, allowing for the generation of a wide variety of raw content through conditional prompting. So users can now access all the sounds they want and need just with the text prompt expanding for the possibilities of audio content creation. All right, so let's go ahead, let's go ahead and take a listen for. I, Hey, I, I always love when our, our live stream audience can, can let me know if you can hear this, but we are going to just listen to maybe about 20 or 30 seconds here for this new 11 labs announcement. So let's go ahead and listen. So for our podcast audience, you're not going to see what's popping up on the screen, but essentially this is reportedly, uh, you know, just text inputs and everything that you're hearing here. And this is a tweet from 11Labs or an X. I mean what do you call that now? I don't know. I know I, I know I'm hard headed. I still call it Twitter or a tweet, but you have an axe tweet from, from open or sorry from 11 labs here showing kind of the prompts and the, the output that it's producing. So presumably everything that you're hearing here is from this new updated model. So let's go ahead and take a listen. Live stream audience, let me know if you can hear this. Here we go. 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, 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 GEni. But now dreams can be heard, not just seen. That's right. Create any sound you can imagine. All right, so hopefully y' all can hear those sound effects kind of going on in the, in the background. But let me just kind of play these, these last couple here. So we go from seagulls calling to a grunting pirate to bouncing gold coins and chicken. All right, well, let's just, let's just listen to just that kind of series here. So we're gonna go from pirate to bouncing coins to chicken. All right, let's go ahead and listen. Here we go. So, I mean, fairly impressive, right? So, you know, the crazy thing is, is I've seen actually, I believe one of the top sound design labs in Chicago, so I'm based in Chicago here is Tribeca. And I've, you know, got to see how they produce a lot of these sound effects and just the amount of people and, and props and technology that's needed to produce very simple sound effects. You really wouldn't think about this, right? But when you're watching a, a blockbuster movie or a TV series, you probably don't see, unless you're involved in the industry, all of the work that goes on behind the scenes for all of those small little sound effects, right? And just, I've seen these giant, giant rooms, right? And they and they said, oh, this is our room just for glass shattering. Right. So this is pretty disruptive technology. Is it studio quality yet? Probably not. Will it probably be there in a couple of months? Yeah, you can make that argument. So, you know, we've seen similar announcements from Meta, Google, some other big players, but eleven Labs has been one of the household leaders in the text to voice space. So actually, hey, I'll, I'll go ahead and tell you. Hey, if you listen to our podcast, there's a little intro and outro that was done with 11 labs. Right? It's, it's been a leader here for years. You know, I, we actually follow this space fairly closely. I'm personally intrigued by it. So I like what 11 Labs is doing here. So in the same, in the same way that you are seeing, you know, chat GPT, you know, go kind of multimodal, right? So you have this, this new data analysis, the V2 of data analysis, you have the ability to, you know, create images. I'm sure at some point we're going to have the ability to create videos. So I mean, here you have 11 labs doing something similar, right? So it's like, yes, you have the ability to create voices, but now you have the ability to create soundscapes and sound effects. Right. And presumably they're probably gonna have, you know, audio and music at some point. So, you know, Suno is, is kind of the leader in that. Suno and Udo. So pretty, pretty interesting move here. So, so promise here on the live stream says that, uh, I'm amazed. Uh, Jay. What's up, Jay? Jay says, very impressive. So, uh, yeah, I think it's pretty, pretty big, right? It doesn't seem very big. But think of all the different applications. Even if you're, you know, small, medium sized business, maybe you're not in, you know, animation or something like that, but maybe you have a YouTube series or you're trying to create, you know, marketing videos that, you know, resonate a little more. I mean, just right there, just having the kind of background soundscapes and the ability to kind of have different sound effects. Pretty, pretty impressive, especially with just text. All right, let's keep going. A couple other pieces of AI news that matters. Yeah, yeah. Love what Jennifer says here. She says, well, the glass shattering room can always be used for an extra break room. Yeah. Hey, humans, humans are going to get tired from producing so much generative AI content with text. Right? All right, let's, let's go on to our next piece of AI news. And if you follow US daily. We covered this midweek, but this is. Hey, if you're just joining us once a week, we gotta talk about this. So Apple is not just in talks. We're going to say they're pretty close to finalizing their kind of deal with OpenAI to enhance Siri with OpenAI's Chat GPT technology. So according to Bloomberg reports, Apple has been negotiating with OpenAI since mid 2023 to integrate Chat GPT technology into Siri, aiming to improve its responsiveness and natural language processing. Capab. Apple engineers have been testing ChatGPT APIs with Siri, demonstrating the ability to handle more complex queries and understand user context better. Also, Microsoft is reportedly a bit concerned about this new partnership between Apple and OpenAI as it relies on OpenAI technologies to power Microsoft Copilot. And we just saw at the Microsoft Build conference all these future capabilities and functionalities within its Copilot system. And this is reportedly raised a lot of worries internally at Microsoft. So, you know, Microsoft has reportedly invested between 10 to $13 billion in OpenAI for a 49% equity stake. So, you know, kind of some reports are telling us that Microsoft is a little concerned now that OpenAI is essentially getting in bed with its closest competitor. Right in. In Apple. Right. So pretty interesting here. We've been hearing about this off and on for the last six months, what Apple is going to do. So, you know, originally the first reporting was, oh, they're going to be partnering with OpenAI. And then it said, oh, nope, they're actually going to be developing their own model. And we kind of saw some research papers if you tune in here every Monday. We've been talking about some models that Apple has actually released. And then it was reported, oh no, it's actually going to be Google and Gemini coming to future, you know, Apple devices. And you know, at least over the past month, everything else has been squashed. And we are looking at, presumably what is actually going to happen here is this Open AI and Apple partnership. So this is especially timely now and we'll be talking about this definitely next week, because that is happening next week. So Apple is set to unveil all of their new AI features next week, presumably powered by OpenAI at their WWDC conference, which happens one week from today, so June 10th. So this is including, so what we've kind of seen from reporting, mainly from Bloomberg, including real time translation, enhanced Siri capabilities. That's the big one, possibly even labeling them as a beta preview to signal ongoing development. So with this new partnership with OpenAI Siri will reportedly get much smarter capabilities similar to Those demonstrated by OpenAI in their GPT4O model, which we talked about here a couple of weeks ago. Also, you could be seeing some new agent ask capabilities with this new feature update, including the ability to perform actions inside of Apple or iOS apps with just your voice. To me, that sounds a little counterintuitive, except for accessibility reasons. You know, maybe if people, you know, aren't able to use their hands for certain reasons, I can see some benefit there. But otherwise, I don't know. For me, I'm not going to be talking to Siri to try to control, you know, apps on my screen. I personally think that'll be a little clumsy, but there are some huge accessibility benefits there. So that's, that's pretty cool. So, like we talked about, this new WWDC or, sorry, The WWDC announcements June 10th are going to be showcasing the first preview of the new iOS 18 AI features, and we'll finally, in one week, get final confirmation on what these new Apple advancements are going to be. One new piece of reporting is we are unlikely, at least as of now, to see any new Apple hardware, which would be interesting. So we might see announcements on that maybe in September. So kind of my, my big takeaway from all of this is, you know, Apple, I've said this all along. Apple's always the latest to the party, but they're generally the coolest kid there. And some people have asked me, like, hey, Jordan, do you think that this, you know, Apple and OpenAI partnership is going to be better than the Microsoft and OpenAI partnership? And I'd say probably not. Right. Full disclosure, I'm coming to you. My video is on my iPhone, right. I'm on my Mac, right. So I've been kind of Team Apple or Team Mac for like 10 to 15 years, but I'm thinking about switching. Right. I've been personally impressed with everything that Microsoft has been developing. You know, at their Microsoft Build conference, you saw their, the recall feature, which, the ability on certain Microsoft Copilot Plus PCs, it literally just records everything you do on your PC and then you can go back at any point and talk to it conversationally. So when you see what Microsoft is, is going to be releasing, and that presumably is going to be June 18th that we start to see some of those features released in the wild with the new copilot plus PCs, I don't know, at least for me, I think whatever Apple comes up with, presumably we're going to see it first in the newest iPhone in the iOS 18 operating system. Although reportedly a lot of these features are only going to be available in some of the newest iPhones. But still, I think Microsoft has had a two plus year head start dating back to its partnership with OpenAI. So I feel, you know, even though recent reports from Bloomberg say that they've been negotiating with OpenAI since mid 2023, I mean, Microsoft has, you know, presumably had tens of millions or hundreds of millions of devices that are reporting back right, with their current partnership. So Copilot has been powered by OpenAI's GPT technology all along. So I think Apple's gonna just come pretty late to the game, right? And I think that they're gonna have to come up with, I'm not gonna say gimmicky, but kind of gimmicky features, right? Like we've heard all this, like, oh, there's gonna be some, you know, AI powered emojis. It's like, okay, does that really, I mean, that's a great little marketing piece and it'll probably, you know, probably be entertaining or amusing for a couple of minutes. But if that's one of the marquee features that we're hearing from Apple, I don't know if I'm in, you know, if I'm a higher up at Apple, I'm not feeling too great about its position right now just because Microsoft is so far behind. However, I do think ultimately this is the best move for Apple, right? So I even said initially when Apple was, you know, reportedly spending millions with an S. Millions of dollars a day on its own model, and then, you know, their report saying that they were going to be using the Google Gemini model, I said, good luck. I said, if they go down either of those two paths, and I said this like, I don't know, eight months ago. I said, if they go down either of the those two paths, they are screwed. They are screwed because OpenAI is so far ahead of everyone else, y'. All. I talk about large language models every day. I spent countless hours testing large language models every single day. Right now OpenAI's model is light years ahead of everyone else. I don't care. Google Gemini 1.5 Pro, advanced, flash, whatever model you want to call it, Claude 3 opus. I mean, OpenAI is so far ahead, right? And they just gave us a little teaser with this GPT4.0 model, which is not going to be, you know, its most powerful model of the year. We're going to see a new model released. You know, whether that's called, you know, GPT5, whether that's called GPT next, which is another potential naming that, that we've heard. But regardless, OpenAI is light years ahead of everyone else. Right. We're still working with, you know, an updated version of GPT4 which is was trained two years ago. Right. And you have all these other companies, you know, releasing models today that are so much newer and they still cannot out out benchmark or out. In my opinion, they cannot even be in the same conversation with OpenAI right now. So overall, smart move from Apple ultimately reportedly picking OpenAI for its future of Edge AI on device, AI with iPhones, with Macs, you know, presumably will be whether we're seeing it at this June 10th WWDC announcement or not, whether we are seeing it come to its kind of computer line in max or not, at least I think they made the right choice. But I think Apple is so is going to be so far behind Microsoft. If Microsoft could get their marketing and positioning right, I think that Apple could be in trouble. But Apple are, they are masters at marketing, masters at advertising. So I mean, we'll see how it all shakes out. All right. And then last but not least, our last piece of AI news that matters. Well, there's more OpenAI drama after a new podcast appearance spurred a bunch of mud kicking, more or less. So Helen toner, a former OpenAI board member, went on the TED AI podcast to discuss all things OpenAI and all the drama that happened with Sam Altman back in the fall. So Toner accused the former CEO Sam Altman. Sorry, accused CEO Sam Altman of lying to and obstructing the board, creating a toxic atmosphere and retaliating against critics. So on the podcast, Toner also said that OpenAI that the OpenAI board found out about Chad GPT's release on Twitter of all places, which I found that one interesting. I don't know if you're the board member, if you're on the OpenAI board, are you really going to find out about its marquee product being released on Twitter? That to me, you know, got me scratching my head. So if you don't know any of all the, all the drama, right. So yeah, now all of us geeks have the equivalent of, you know, I don't know, Bravo, Real Housewives or something, right? We can follow all this, you know, tech, tech drama. But. So Altman was fired on November 17 last year after weeks of intense discussion within the board due to quote unquote, long standing patterns of behavior. So Altman was then after being fired on Nov. 17. He was rehired on Nov. 20. So Toner said examples of Altman's alleged behavior include not informing the board about the release of ChatGPT and misleading the board about his financial interest in the company. Juicy, right? So Altman reportedly, according to Toner, targeted her after she co authored a critical research paper leading to further distrust within the board. Toner also detailed and what she said, OpenAI employees just supported Altman's reinstatement out of fear of retaliation and a false dichotomy that the company would be destroyed without him. And in the days since this interview, both Sam Altman and other board members have responded to Toner's accusations. So we'll be including those in the newsletter today as well. So if you want to see how the OpenAI board responded and how Sam Altman himself responded after this interview with former board member Helen Toner, make sure to check out the newsletter today. So yes, yeah, I love this. Yeah, Joe said. The AI drama version of Bridgerton. Yeah, so much, so much drama. So let's go ahead. Let's quickly recap all the AI news. Well, not all the AI news that matters, but the AI news that matters that we went over today. So Nvidia unveiled its next generation AI chip architecture Ruben after just three months after unveiling Blackwell and also confirming a yearly rollout cycle. Next piece of AI news is every single big company except for Nvidia has essentially teamed together. So Google, Microsoft, Meta, amd, Intel and others have joined it to form the UA Link group to compete with Nvidia's NV link technology. More chips. So also at Computex, AMD announced new AI chips to try to compete with Nvidia. We'll see if that does anything popular. AI text to speech company eleven Labs is now releasing text to sound which we heard so you know, with soundscapes, instrumental tracks, character voices and sound effects. And then Apple, we should be hearing its final kind of next week. It's its final partnership with OpenAI, but we've heard and we detailed some of the rumored capabilities of its new partnership with OpenAI and we'll hear about that at WWDC one week from today on June 10th. And then last but not least, detailing more open AA drama after former board member Helen Toner went on a podcast and then Sam Altman and other board members responded. So I hope this was helpful. Jay said it was. If this was helpful, let me know. We do this every single Monday. So if you are new here, thank you for listening on the podcast. If this was helpful. We'd appreciate it if you could leave us a rating on Spotify or Apple. Tell your friends subscribe to the show. We do this every weekday, so normally we have guests. So. But every single Monday we bring you the AI news that matters. So maybe you don't have, you know, 25 minutes every single day to join the show, but you want to keep up. So Mondays is then the day for you and if you are on the live stream, hey, like Fred here holding it down with me in Chicago. If this was helpful, go ahead, tag a colleague in the comments here on LinkedIn. Click that repost. We spend hours every single day sifting through all of the important things happening in the AI space. If this is helpful, please go ahead and click that repost button. It takes you three seconds. Takes us many hours every day to bring this to you. So thank you for tuning in. If you haven't, go to your everyday AI dot com, sign up for that free daily newsletter and we'll 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 everyday AI.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.