Ep 300: NVIDIA becomes most valuable U.S. company, Runway Gen-3 looks impressive – AI News That Matters

Resources

Join the discussion: Ask Jordan questions on AI

Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup

Connect with Jordan Wilson: LinkedIn Profile

Try a Free AI Course: Register for our free Prime, Prompt, and Polish AI Course! 


A Competition of Titans: OpenAI, Meta and Anthropic

The world of Artificial Intelligence (AI) is flourishing, with companies such as OpenAI, Meta, and Anthropic leading the frontline. One of the significant developments is OpenAI’s possible release of a new language model. Meanwhile, Anthropic has made a watershed move with its Claude 3.5 SONNET that features 'artifacts,' enabling real-time rendering of code and other file types.

NVIDIA on the Rise

Besides AI development, the tech world witnesses NVIDIA's market cap skyrocket, temporarily surpassing titans Microsoft and Apple. This escalation is primarily attributed to NVIDIA's supremacy in providing GPU chips for AI, predicting a race towards a $4 trillion market cap in tech among these stakeholders.

A landmark moment was achieved when NVIDIA CEO participated in a first-of-its-kind live AI-to-human interview facilitated by the partnership with Hour 1. Simultaneously, OpenAI has entered strategic alliances for cancer research with a healthcare start-up, Color Health, and infrastructure enhancement through the acquisition of RockSet.


OpenAI Safety Concerns

OpenAI has faced criticism, being accused of overlooking safety in favor of offering product updates and procuring partnerships. On the other hand, OpenAI's former co-founder has embarked on a new venture, SAFE Super Intelligence Inc., placing a prominent emphasis on AI safety measures.


Accelerated Developments

In the same vein, Anthropic showed its prowess by unveiling the Quad 3.5 Sonnet, benchmarking as the most capable model currently, thereby staking a claim in the AI sector. Meanwhile, Microsoft and Apple are locked in a close competition for being the most valuable company, with NVIDIA also making its presence felt significantly.


Media Transformed

On the multimedia front, Runway introduced its Gen 3 Alpha, an AI video entrance with promising faster generation times. Similarly, Anthropic’s Claude 3.5 SONNET is creating ripples in the AI chatbot arena with its improved functionality and efficiency for diverse tasks.


Broadening Horizons

OpenAI's collaboration with Color Health aims to revolutionize cancer screening and treatments, leveraging AI models. Secondly, the Rockset acquisition is expected to gear up OpenAI's ability to analyze vast data amounts for enterprise products rapidly.


The AI Future and You

In the whirlwind of innovation and competition, the AI industry’s ongoing evolution promises an advanced range of solutions to the market. As a business owner or decision-maker, staying updated on these developments will be critical to making informed decisions about leveraging AI technologies. Immersed in this exciting transformation and armed with insights from this sector, there are immense, untapped possibilities to harness for future growth.

Topics Covered in This Episode

1. Competition in development of LLMs
2. NVIDIA's rise as the most valuable company.
3. OpenAI drama and partnership
4. Runway's new AI model


Podcast Transcript

Jordan Wilson [00:00:16]:
Is chatGPT in trouble? Well, anthropic Claude just dropped 3.5. SONNET with artifacts. Speaking of dropping, well, NVIDIA dropped after becoming the most valuable company in the world. They aren't anymore. And an OpenAI cofounder is starting a new company that I think is going to be pretty impactful. Yeah. As always, so much going on in the world of AI news. You can spend hours a day trying to keep up and trying to really understand what it means, for you and growing your company and growing your career, or you could just tune in here every Monday where we bring you the AI news that matters.

Jordan Wilson [00:01:02]:
What's going on y'all? My name is Jordan Wilson, and this is Everyday AI. Thank you for watching. Thank you for tuning in, and this is for you. It's your daily livestream, podcast, and free daily newsletter helping us all learn and leverage generative AI to grow your company and grow your career. And one of the hardest parts about doing just that is keeping up with everything that's going on. Right? Because you're trying to make long term plans for your company, and you see all this stuff that's going on. And you're like, well, what does it actually mean? Well, that's what we do every single Monday is we tell you what it actually means. So, if you're joining us on the podcast, check out your show notes, a link to our website there and a lot more.

Jordan Wilson [00:01:40]:
So go to your everydayai.com. Sign up for the free daily newsletter. So, yes, we do the news on Monday, but we do this every single weekday. Normally, we're bringing on expert guests. And then in our newsletter, we always recap, the show for each day as well as share about news every day, fresh finds from across the Internet, and just what you need to know, to get ahead and stay ahead with generative AI. Alright. So without further ado, let's jump into it and go over the AI news that matters for the week of June 24th. A lot going on.

Jordan Wilson [00:02:17]:
Let's jump right into it. So Runway has struck back with its new gen 3 alpha in the AI gener video generation wars. Yeah. It's been crazy. Alright. So Runway has introduced gen 3 alt alpha, a significant advancement in generative AI video creation following the successes of its gen 1 and gen 2 models. So, this new model, not yet publicly available, but it promises high quality, detailed, and highly realistic 10 second video clips with a variety of emotional expressions and camera movements. Alright.

Jordan Wilson [00:02:53]:
So, Runway has obviously been one of the leaders in the AI video space, but their new product here, Gen 3 Alpha, boasts faster generation times with a 5 second clip taking 45 seconds and a 10 second clip taking 90 seconds to generate. So we'll see what those generation times are once it's actually rolled out because once these new updates come, you know, they're gonna get slammed. Right? So, the minute or the day or the hour that this drops, if you go on there, yeah, it's gonna take a lot longer than that. But the model will initially be available to just paid subscribers with plans to extend access to free users in the future, which is pretty, pretty important there. Runway's cofounder confirmed that the model will be released to paying subscribers in a matter of days. So, we heard this. It's been a matter of days now, so we'll see when that actually happens. But Runway's gen 3 alpha represents a pretty big step forward toward, building what they call general world models, which aims to simulate a wide range of re of real world situations and interactions.

Jordan Wilson [00:03:59]:
So, yeah, the last couple of weeks, if I'm being honest, the last couple of weeks here in, June 2024 and late May, I mean, we've seen more action in this AI video space than the previous, like, 5 months. Right? So we saw the Cling, AI video gen generation product, from the Chinese company. Right now, that's not available to anyone unless you have a Chinese phone number and you can verify that. But then we saw a dream machine. Right? We saw a dream machine from Luma Labs, released publicly to everyone. That was about a week ago. And, you you know, Sora, we are still waiting from Sora from OpenAI. So, you know, I talked about this on the show over the last couple of weeks that it seems like run Runway and Pico Labs might be falling behind with some of these new competitors.

Jordan Wilson [00:04:46]:
Right? So obviously, OpenAI's SoRA is not out yet, which did seem just light years from a quality standpoint ahead of everyone else. Right? So OpenAI has made Sora available, to a lot of, you know, higher end content creators, to, red teamers, to some very visible people, and then to people like, oh, Ashton Kutcher has access. Right? But the generations we've seen from Sora, I think, put it far ahead, not far ahead, but, at least decidably, decidedly ahead of everyone else. But then I think you have, Cling kind of in there in the conversation, and now we're gonna be looking at runway, in this new gen 3 alpha. And if this is going to, kind of jump over the upstart, you know, startup, drain machine from from Luma Labs. So, we'll see where this, goes once it's available. But, I mean, initial initial generations look pretty impressive. So make sure to check out, the newsletter today because we will be sharing, some of those generations so you can see.

Jordan Wilson [00:05:52]:
Alright. Gotta get that a little little drink of coffee. And, alright, Speaking of, competitor products and some pretty exciting releases, well, hey. Big news. Anthropic just unveiled Claude 3.5 SONNET, a pretty big breakthrough, in AI chatbot technology. So Anthropic, a key player in the AI industry, has introduced Claude 3.5 SONNET, its latest and most powerful artificial intelligence model. So Claude 3.5 SONNET boasts enhanced capabilities and understanding nuances, humor, and complex instructions, making it proficient in generating high quality content with a natural tone. The new model is much faster than its predecessor.

Jordan Wilson [00:06:40]:
Kind of the most powerful model previously was Claude 3 Opus, more on that in a second, showcasing the company's commitment to innovation and advancements in the AI chatbot space. So the new Claude 35 Sonnet is available for free on the Claude dotai website and the Claude iPhone app catering to a wide audience interested in leveraging this new cutting edge AI technology. What I think is the attention grabber here is the introduction of what they're calling artifacts, which further enhances user experiences by enabling tasks such as generating text documents or code and providing dynamic workspace for real time editing and collaboration. So this development signifies a significant step forward in the AI technology, offering users improved functionality and efficiency in various tasks from content creation to code execution. And, obviously, what we gotta talk about now is the unveiling of Cloud 3 5. Sonic reflects the ongoing evolution and competition in the AI industries with companies like Anthropic, OpenAI, Google, you know, continually pushing the boundaries to deliver more advanced solutions to the market. So let's talk a little bit here, about the strategy in some of these new features. Right? So, kind of the strategy side.

Jordan Wilson [00:07:57]:
So previously when we got, when we got Claude 3, we had Claude 3 Haiku. And this is in order from kind of the, quote unquote least powerful to most powerful. So So we got Haiku, we got Sonnet, and then we got Opus. So Opus was technically their most powerful model, so a lot of people were wondering this 35 version only came to the middle model, only came to SONET. And a lot of people are are wondering why. I have some, brief takeaways on that. But first, I wanna talk about the benchmarks. So, what's what's important to always keep in mind, whenever a new company, you know, releases a new model, they always bring with it their own benchmarks.

Jordan Wilson [00:08:39]:
Right? What's important to keep in mind is these benchmarks are always kind of by the company. You know, generally, they are confirmed by 3rd parties, but just keep that in mind. But from a benchmarks standpoint, this kind of quote, unquote middle model, but the updated version 3 5 Sonnet, is already benchmarking ahead of the kind of newer model, from OpenAI, their GPT 4 o, right, which was just released a couple, technically like a month and a half ago now. Right? So a lot of just back and forth. It's a slugfest. And and poor, you know, poor Google Gemini in all this doesn't really get any of the headlines. You know, it's curious how far behind Google seemingly is. And it's really just been, you know, OpenAI and Anthropic just trading blows back and forth in terms of who is going to have the most powerful model.

Jordan Wilson [00:09:31]:
Right? And we're gonna be talking more about this later this week, you know, and talking about some of these benchmarks, the MMLU benchmark, also, like, kind of the release cycle here. Very interesting. Right? So, essentially, when Claude 3 came out, they were kind of the, quote, unquote, top dog for just a couple of weeks. Until then, OpenAI released their updated model. You know, they kind of came out with the GPT 4 turbo, and now we have the GPT 4. Oh, so So it's just been a slugfest back and forth, and everyone wants to be, kind of the highest, highest rated in these benchmarks, the most capable models. So, the interesting strategy here that we see from Anthropic, is updating the middle model. And you might be wondering why.

Jordan Wilson [00:10:17]:
You know? Similarly, when we saw Meta, you know, Facebook, parent company there, they released their, pretty much open source MADA, model, llama 3. But they didn't, release yet their biggest version, the 400,000,000,000 parameter version. So it seems like kind it's it's a cat and mouse game right now because I think it is widely known in the industry that whenever OpenAI does unleash, whether it's called GPT 5, GPT Next, you've seen different, kind of names floated out there. Maybe it'll just be GPT 4.5. We're not sure. But it seems like these other companies are kind of holding back in waiting for this next open AI real OpenAI release. They kind of see, okay. You know, we can release this this middle model or in meta's case.

Jordan Wilson [00:11:06]:
Right? Meta's, you know, middle model did very well. Right? From a just from a size, it is, you know, might be a champion of its weight class, so to speak. But, pretty pretty interesting developments and a pretty interesting strategy here, from Anthropic with their new Quad 3 5 SONNET just by updating the middle model. The artifacts feature, fantastic. We did a a pretty quick review couple hours after it came out on our YouTube channel. Yeah. It seems seems sounds weird talking about the YouTube channel, but pretty good, you know, pretty good comparison. We'll try to throw that in the show notes or you can just go check it out, find everyday AI on YouTube there.

Jordan Wilson [00:11:44]:
But the artifacts feature, I think, more so than the benchmarks. Right? Yeah. You know, the benchmarks are gonna come and go. Right? The benchmarks, you know, could be gone next week if OpenAI decides to, you know, release its new model, and they'll probably be crushing everyone else in benchmarks. I do assume whenever OpenAI releases their next model, they're gonna be so far ahead. It's not going to be close. Anyways, the artifacts feature, though, I don't think is getting a lot of shine. Right? So, essentially, you have this kind of dual panel.

Jordan Wilson [00:12:14]:
And on the right hand side, you essentially get a live preview. So let's say you're doing some code generation. Right? You know, generally what you would have to do is copy and paste that code. You'd have to go into some kind of environment, maybe your website or, you know, a site like Replit, you know, where you can just sign up for a free account and essentially, you know, deploy sandbox instances to see, oh, does this code work? So what I love from this artifacts feature from Anthropic is it does it right there for you. Right? It literally renders the code that you're working on. It can render, you know, different types of files, SVGs, interactive dashboards. Right? Like this changes, I think, how largely how large language models can be used. I think they're actually, I'm being honest, I think they're actually borrowing from the open AI playbook because we actually first saw this in the GPT builder, right, from open AI.

Jordan Wilson [00:13:07]:
And now that you have the ability to build a GPT kind of conversationally, on the right hand panel. So you have this in OpenAI when you're building a GPT. You can kind of build it on the left hand side. That's just a custom version of ChatGPT. If you don't know, you can upload your own, knowledge base, etcetera. And then on the right hand side, it renders out. So I I I think they're taking a page, from OpenAI, but, hey, they're the first ones to actually bring it like this. Right? To be able to see how this, you know, graph might look, how this web design might work, how this, other design might work, you know, to to be able to render it in real time, I think is a pretty game changing feature from Anthropic.

Jordan Wilson [00:13:48]:
I've been kind of hard on Anthropic just because, if I'm being honest, I don't use it a lot. Right? I've paid for it since day 1. I always go in there, I would say, at least, at least a couple of times a week. But the fact that it is not connected in real time to the Internet, I think, is a huge, disservice. But, we're gonna have more on that later, actually, tomorrow. We're gonna be comparing Infropic's new Claude 3.5 SONET and how it compares to ChatGPT's 4 o. So we're gonna have more on that tomorrow. Don't wanna give it all away, but make sure to check out that show.

Jordan Wilson [00:14:19]:
Alright. Our next piece of AI news, well, we have tech titans just jostling for the title of the top company in America and, well, the world. So, NVIDIA, last week, their share price soared to an all time high, giving it a market capitalization of $3,340,000,000,000,000, which is nearly doubling its market cap since the beginning of the year. That's wild to think about. Alright. So, with that, NVIDIA briefly held the title of the world's most valuable company by market cap. So the company's, NVIDIA's growth has been fueled by its straight up dominance in providing GPU chips for artificial intelligence, which a lot of people are considering, you know, the new gold or oil or data, like right? Whatever the most valuable, thing in the world is. Right now, it is GPUs, it seems like, because that's what is, you know, kind of launching companies into being the most valuable in the world.

Jordan Wilson [00:15:16]:
However, NVIDIA's shares slumped Friday, and Microsoft regained the top spot. So Microsoft reclaimed the top spot as NVIDIA's shares fell by about 4%, reducing their market cap now to about 3.1 trillion. So, yeah, pretty pretty bad day for, NVIDIA to lose 200 $1,000,000,000 in market cap in a single day on Monday, but, you know, this stuff goes back and forth. So we've seen just over the last week. I don't know the last time that this has happened that there has been 3 companies in Microsoft, Apple, and NVIDIA that are constantly going back and forth for the kind of the, quote, unquote, title as the most valuable company in America. It's been interesting to watch, sometimes changing multiple times a day. So kind of analysts right now are predicting a race to $4,000,000,000,000 market cap in tech between NVIDIA, Apple, and Microsoft. So pretty meteoric rise, obviously, out of NVIDIA, whereas, you know, 3 years ago, unless you were maybe a a a Bitcoin miner or a gamer or you really just followed semiconductor technology, most people in the world had never heard of NVIDIA.

Jordan Wilson [00:16:33]:
Right? Whereas other, you know, kind of big name companies like, you know, Microsoft, Apple, you know, Google, Amazon, everyone else has been a household name. So, you know, NVIDIA has literally come out of nowhere for the most part over the last year and a half, 2 years to now being a world superpower, which obviously we told you about a year ago, on this show. We said no one's paying attention to NVIDIA. You need to be paying attention. We called it back then. We said they're gonna become the most important company probably in the world, and we've already seen that now. Alright. Speaking of hey.

Jordan Wilson [00:17:04]:
If you if you're on the, podcast have on our screen here kind of a photo of us, NVIDIA CEO Jensen Huang. They're talking about the, NVIDIA news. You know? Do you know that he actually had a a clone made of himself by, the company Hour 1? So, we can, hey, we can talk about did we make news? Maybe. I think we actually did. So, hey, it's it's our show. I can report on this. So speaking of some kind of news here, well, yeah, we kind of made some news, I think. So we partnered with our one last week, and I believe we became the 1st ever to do a live AI clone to human interview.

Jordan Wilson [00:17:46]:
Yeah. I think that's a thing. But, yeah, we did it. So we did partner with PowerOne last week to create a super realistic AI clone of myself, and then my clone interviewed me. Yeah. Kind of kind of meta. Right? I'm here telling you about last week when my clone interviewed me. It like, anyways, I kinda think I got my butt kicked.

Jordan Wilson [00:18:07]:
I don't know. Our our livestream audiences, you know, if if you caught that one last week, let me know. I think actually my AI clone got the better of me. But a couple kind of big names, have done this as well. So like I said, NVIDIA CEO Jensen Huang, worked with Hour 1, the company to do this. The same thing with LinkedIn cofounder, Reid Hoffman. So, you know, let me know if if the 3 of our clones should maybe do a show together. Me, the LinkedIn cofounder Reid Hoffman, and, NVIDIA CEO Jensen Huang.

Jordan Wilson [00:18:39]:
I kinda I kinda think we should get that going, but let me know in the comments if that would be a fun one. The world of communication is changing. That's why you need to pay attention to a sponsor of this podcast, hour 1. Even NVIDIA CEO Jensen Huang used hour 1, and I think you should too. So my name is Jordan Wilson. I'm the host of Everyday AI, and here's the deal with hour 1. You gotta check them out. It's a time saver for communicating at scale in ways you literally couldn't before.

Jordan Wilson [00:19:11]:
So let's say you need to deliver a presentation to global audiences in different languages you don't speak, or maybe you deliver live trainings every single week that are 95% the same, and you're only changing a few things. That's where hour 1 comes in. With a host of features like AI avatars, premade and custom templates, and collaboration tools, Hour 1 is an all in one AI video platform to communicate at scale in new ways that weren't possible before. So make sure to check out the show notes or go to your everydayai.com for info on how you can actually win a free custom AI avatar from Hour 1. Go check it out y'all. Alright. Our next piece of AI news. Yeah.

Jordan Wilson [00:19:57]:
Just a lot of OpenAI related news here. So, OpenAI has partnered with Color Health to enhance cancer screening and treatments using AI models. So Color Health's AI copilot, which is powered by the GPT 4 o model, assists doctors in creating personalized cancer screening and pretreatment plans. So the copilot integrates with patient data, clinical guidelines, and AI technology to optimize diagnostic tests and streamline the pretreatment process. So early studies are showing that a month's delay in cancer treatment can increase mortality by 6 to 13%, highlighting the critical role of efficient AI assisted processes. So this new collaboration aims to alleviate administrative burdens on oncologists, allowing them to focus on critical decision making in cancer care when treating, cancer patients. Alright. More OpenAI news.

Jordan Wilson [00:20:56]:
Yeah. It's been a lot of it this week. So this one, I think, is pretty big. I don't think enough people are talking about this one. So OpenAI has acquired Rockset. Alright. So, Rockset is a search, kind of a a database analytics, startup. So this, acquisition here came in a stock deal reportedly valued at a few $100,000,000.

Jordan Wilson [00:21:22]:
So Rockset is known for its real time search and analytics databases and will bring expertise in real time data processing and vector search to enhance OpenAI's enterprise products. So the acquisition aims to improve OpenAI's ability to quickly assess and analyze vast amounts of information, leading to faster and more accurate responses from AI models. So Rockset's team will join OpenAI powering the retrieval infrastructure of the ChatGPT maker's enterprise products. OpenAI's move to acquire roxaet signifies its commitment to expanding offerings and staying competitive in the AI market, particularly in the enterprise sector. So the integration of RockSets technology is expected to benefit benefits enterprises by providing tools to rapidly index and search through their data, potentially improving operational efficiency and decision making process. I think this is a huge piece of news that hardly no one is talking about. I think maybe one of the reasons why is this does just interact sorry. This does just kind of affect, enterprise customers.

Jordan Wilson [00:22:34]:
Right? And obviously OpenAI and ChatJPT with 100 of millions of customers, so many of them are consumer. Right? But this is huge because one of the the knocks on all large language models. Right? Not just ChatGPT. Maybe in this space actually Gemini is a little ahead of everyone else, but, you know, same thing with anthropic law. So one of the biggest things as well, you know, hey. These large language models are great and we see the capabilities of them. However, what about our company's data? Right? Especially an enterprise company. That's what makes it difficult.

Jordan Wilson [00:23:06]:
Right? Because when you look at these models, you have what's generally called a a training date cutoff or a knowledge cutoff. Right? And sometimes that is it can go up to 6, 8, 9 months. Right? Depending on how quickly these companies are updating their models. So that means, you know, let's say you're an enterprise company. First of all, you don't necessarily know if the information that these big large language models have scraped is the most up to date and accurate information even as of that cutoff date. So that is one of the biggest problems, and it does actually cause hallucinations. Right? Because if a model isn't a 100% sure if you're asking it a very specific query, that's one of the reasons why sometimes you get these vague cookie cutter responses. Right? So this move from OpenAI, I think is pretty huge, and it will impact, some enterprise customers.

Jordan Wilson [00:23:58]:
So we are actually doing a, a training for an enterprise company last week. And this is one of the most common, you know, train, questions that we get in prompt engineering in general, especially when we're helping companies. Right? That's something we do. So if that's something you need help with, make sure to reach out to us. However, you know, companies are like, wait, you know, even if we have the enterprise version of ChatGPT, it's not connected to our our data. Right? It's it's it's our workspace. You know, we can create GPTs with our own data. It's like, no.

Jordan Wilson [00:24:26]:
It it's your own data is not there. It's not live. So I think this is a pretty big play in that space, with OpenAI's acquisition of RockSets, trying to you know, I don't know if this will, in theory, solve, quote unquote retrieval augmented generation. So we talk about bringing rag, to models. Right? So you could bring your own kind of data, but right now you have to use a a third party. Right? You have to build a solution around this and you can't really do this out of the box inside the kind of quote unquote chat interface of ChatGPT. You have to build this. You have to be able to work with, 3rd party tools, systems, integrations, APIs, you know, your data warehouses, etcetera.

Jordan Wilson [00:25:05]:
So it's gonna be interesting to see how OpenAI uses, this RockSett, kind of acquisition to maybe change that. I don't know if it'll solve it, but it's going to change it, I think, for sure. Alright. And our last piece of AI news for the day or might be for the week, but, again, huge. This is huge. I don't think, you know, people are really paying enough attention to this either. But the former cofounder of OpenAI, Ilya Sutskeber, has launched a new venture called Super Super Intelligence Inc, focusing on developing safe and powerful, artificial intelligence to rival kind of his former company or at least what they're trying to do in this safety sector. So Ilia's departure from OpenAI and the subsea, the subsequent launch of SSI.

Jordan Wilson [00:25:58]:
So, again, that's SAFE Superintelligence Inc. It highlights the increasing concerns in the tech industry regarding the safe and responsible development of AI technologies. So SSI aims to prioritize safety in AI advancements in contrast to concerns raised about prioritizing commercial growth over long term safety at OpenAI. So, obviously, if you follow, the, the everyday AI show here, you know there's been a lot of kind of safety shakeups or deprioritization, at OpenAI. It's kind of their internal team has kind of been dismantled, their safety team. So, the launch of SSI reflects a shift toward prioritizing safety in progress without being influenced by short term commercial pressures. So that's, you know, kind of some of when these, safety, team leads at OpenAI have kind of left over the last month or 2. It's they kinda said they said, hey.

Jordan Wilson [00:26:58]:
Right now, safety is not a big of a concern as it should be. And they kind of accused OpenAI of prioritizing shiny features and product updates over safety, which is something that OpenAI, you know, I think many years ago before all of these, you know, big partnerships now with Microsoft and partnerships with Apple, seemingly OpenAI, at least publicly, you know, seem to be much more focused on, safety and the future of, you know, how, AI and large language models can benefit humanity. Obviously, they they haven't pivoted completely away from that, but it does seem like at least publicly, that doesn't seem like as much of a priority anymore, which is kind of understandable. Right? The fact that OpenAI, is essentially going to be powering literally almost every single device. Right? Not every single one, but I would say more than 2 thirds of devices, it is gonna be powering. Right? Because, Microsoft's Copilot is powered by GPT 4 0 from OpenAI and all of Apple's, products that they just announced at WWDC. Their future, quote, unquote, Apple Intelligence AI is going to be powered by the GPG 4 o technology as well. So here you have OpenAI, you you know, which 3, 4 years ago hardly anyone had heard of.

Jordan Wilson [00:28:21]:
Now they are literally gonna be powering almost the future of how we work. Obviously, Google has their Gemini products in their, Samsung, Android, devices. But, you know, for the most part when you're talking about, you know, desktop computers, laptops, you know, obviously those 2, it's just Microsoft and, and Apple have the lion's share. Right? So, yeah, if you're talking about smartphones, you know, Google obviously, you know, has a big slice of the pie there. But, hey, for the most part, OpenAI, I get it. I get how maybe their their their focus now is a little bit more on these products. Right? Because you have, now this responsibility to, you know, power the future of how we all work. So I do understand maybe how at least from a front facing perspective, how OpenAI is maybe seemingly focused less on safety.

Jordan Wilson [00:29:12]:
I don't think it's a good thing. I think they should be focusing more on safety, but I understand with, especially the, kind of announcements. Right? So Microsoft at their build conference, just, you know, kind of talked about a huge increase, in their partnership and depth of their partnership with OpenAI, and then, you know, Apple going from 0 to a 100 with OpenAI. So I like this approach from, you know, former OpenAI cofounder and, Ilya Sutskever, who is, you know, widely regarded as one of the brightest minds in artificial intelligence maybe ever, but essential you know, especially of of today's current day. So, I think this is super important with his SSI, this new company. But having a third party, that is really, dedicated, to making sure that these AI systems are safe. So we don't know, yet exactly what SSI will be focusing on, but I think it's important. Right? I think you I I think Google does a pretty good job with safety.

Jordan Wilson [00:30:14]:
You have kind of deep mind which is almost it's separate, you know, separate entity. Microsoft I think does pretty well as well. You know, they have their separate branch now called Microsoft AI. But, you know, OpenAI doesn't. And, you know, OpenAI is is seemingly again, seemingly, looks like safety is being deprioritized. So I think it's important for, kind of a quote, unquote impartial third party to be leading the way on safety. I I think probably out of all of the companies, I would say actually Anthropic, has had the deepest and maybe the best commitment to safety at least with, you know, all of these, you know, tech trillionaire or or, you know, startup unicorns. You know, I think, Anthropic and OpenAI are past the, unicorn status.

Jordan Wilson [00:31:00]:
I don't know what comes after unicorn. I think we have to make up a a word there, you know, once you, you know, when an AI startup, you know, hits a $1,000,000,000, you know, valuation, you get unicorn status. So, you know, now that they have $100,000,000,000 valuation, you know, want to come up with something else, super unicorns. Anyways, you know, I think anthropic at least of the, you know, companies building the AI metals has really taken the lead in safety, but pretty huge news here, from Ilya Zetskeber on the SSI. Alright. That is a hopefully quick ish recap, on the AI news that matters for the week of June 24th. Alright. Let's run it down super quickly.

Jordan Wilson [00:31:41]:
So Runway has struck back in the AI video awards, wars with their gen 3 alpha, which should be released any minute, any day now. Then we have Anthropic. Speaking of big releases, Anthropic unveiled their quad 3.5 Sonnet, which is, already benchmarking as the most capable model. Then we have tech titans just jostling for the title of most valuable company in America and actually the world. So we saw NVIDIA briefly claim that title, but now Microsoft, is back and, Apple in number 2 with NVIDIA as number 3. Hey. It looks like everyday AI, we kinda made news of ourself here with the first ever that we believe couldn't find any evidence anywhere else of a clone interviewing its counterpart live. So, yeah, my AI clone interviewed me.

Jordan Wilson [00:32:35]:
I think I got my butt kicked. Then we have, startup Color Health and OpenAI partnered, for a cancer research pilot. More Open AI news where they acquired RockSets, to enhance infra infrastructure for enterprise, products and customers. And then the former OpenAI cofounder, Ilya Setskever, just announced his new company, SAFE Super Intelligence Inc, to focus on safety. You gotta love that he just skipped AGI, artificial general intelligence, and went straight to superintelligence. Talk about, you you know, working in the future and getting ahead. Love to see it. Alright.

Jordan Wilson [00:33:14]:
Speaking of the future, tomorrow, make sure to join us. Yes. That anthropic Claude 35, I think it's actually super impressive and the one feature that I think OpenAI is going to have to respond to sooner rather than later. So make sure to check that out and make sure if this was helpful, if you're listening on the podcast, please, if you wanna help us out, I hope this this show helps you. Please leave us a rating. Subscribe to the show on Apple Podcasts on Spotify. If you're on the, livestream, appreciate our livestream audience as always. Please click that repost button.

Jordan Wilson [00:33:47]:
Takes you, I don't know, 5 seconds. We spend hours every single day trying to bring you the not just the AI news, but that matters, but, you know, insights from experts that can help you grow your company. So, you know, people all the time say, hey, Jordan. This has been so helpful. How can I help? Well, click that repost button if you're here on LinkedIn or maybe tag someone, that needs to hear this. And then last but not least, please go to your everydayai.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.

Gain Extra Insights With Our Newsletter

Sign up for our newsletter to get more in-depth content on AI