Ep 265: ChatGPT’s search engine, Microsoft’s fear of Google, U.S. newspapers suing LLM makers and more- AI News That Matters

Resources

Join the discussion: Ask Jordan questions on AI

Related Episodes: Ep 181: New York Times vs. OpenAI – The huge AI implications no one is talking about
EP 260: AI News That Matters – April 29th, 2024

Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineup

Connect with Jordan Wilson: LinkedIn Profile

AI News and Developments to Watch

AI is always evolving, and keeping track of the latest developments is an ongoing challenge. But it’s a challenge that’s well worth the effort for those who want to stay at the forefront of innovation. Whether it’s new AI competitors entering the scene, massive acquisitions, or efforts to improve AI safety, here’s the news that matters most in the AI world right now.

Investment in Generative AI: A Game Changer

Large scale enterprises are recognizing the substantial potential of generative AI in bringing about infrastructural improvements. The capacity to create digital twins has been highlighted as a transformative venture which simulates potential infrastructural enhancements. This progression has been backed by colossal investment; to the tune of $300 million from the Biden administration, underscoring its importance in the current technological landscape.

Generative AI: Risks and Rewards

While acknowledging potential threats posed by Artificial Intelligence, it was at the Berkshire Hathaway Shareholders conference that the impact of AI was considered, specifically, the challenge of deep fakes and their potential implications.

Simultaneously, AI’s capabilities are opening up unprecedented avenues in the field of medicine. The use of generative AI in the discovery of unconventional novel drug candidates is pioneering groundbreaking advancements in drug development and potentially transforming medical pathways.

AI vs Copyright Laws: An Unfolding Drama

Recently, several major American print entities have sued OpenAI and Microsoft, citing copyright infringement. The bone of contention is the training of AI chatbots using countless articles without acquiring due permissions or adequate compensation, which the tech firms counter as fair use.

Potential Release of New Search Engine Feature by OpenAI

In recent developments, there is talk of OpenAI working on a new subdomain, potentially featuring a search engine and a new chatbot model. Such an enhancement could potentially revolutionize user interactions, providing a more efficient and live alternative to existing search engines.

Microsoft’s Investment in OpenAI: A Tactical Move?

There is a buzz around Microsoft’s significant investment in OpenAI. The European Union has launched an investigation into this move. Some argue that Microsoft’s intention could be to offset Google’s dominant position in the AI space.

The Future of AI: Google vs OpenAI

The imminent launch of a search engine by OpenAI, a firm in which Microsoft reportedly owns a significant stake, could fundamentally transform how people retrieve and utilize information. OpenAI’s proposed search engine, aiming to work on natural language interactions, is poised to challenge Google’s supremacy in the search market. The timing of this anticipated launch could prove strategic, coinciding with Google’s major IO developer conference.

Topics Covered in This Episode

1. Investments in Generative AI and Digital Twin Technology

2. The Impact and Potential Threats of AI

3. OpenAI, ChatGPT, and the Upcoming Search Engine

4. Legal Issues Surrounding OpenAI and Microsoft

5. The Evolving Semiconductor Landscape

Podcast Transcript

Jordan Wilson [00:00:17]:
Is OpenAI now going straight after Google? Speaking of Google, how did Google actually lead to the Microsoft and OpenAI partnership? And what’s going on with all of these US newspapers suing large language model makers? Alright. We’re gonna be talking about that today and more on everyday AI. What’s going on y’all? My name is Jordan Wilson. I’m the host of everyday AI, and this is your guide, to learning and leveraging generative AI to grow your company and to grow your career. So thank you for joining us. If you’re on the podcast, we appreciate you listening. If you could leave us a a review, tell your friends whether you’re listening on Spotify or Apple Podcasts, or or if you’re joining us live, thanks for doing that. So, Rolando and and Dutch and Harvey and Brian and Juan and everyone else, thanks for tuning in.

Jordan Wilson [00:01:04]:
So every single Monday, we bring you the AI news that matters. So you could spend literally hours every single day trying to keep up with all of this, all of these new developments, all these new tools, you know, all these new softwares and how they impact your career and your company. Could spend hours a day or you could tune in every single Monday, to everyday AI where we break it all down for you. We spend time doing this every single day anyways. So on Mondays, we usually give you a recap and talk about why this matters or maybe why it doesn’t. So as a reminder, before we get started, if you haven’t already, this is just the beginning. This is the appetizer. The main course is our newsletter.

Jordan Wilson [00:01:42]:
So make sure to go to your everyday ai.com and sign up for that free daily newsletter for more on today’s stories and a lot more that we couldn’t even, fit into today’s show. So if you haven’t already done that, make sure to do that. And without further ado, let’s get into what’s going on in the world of AI news for the week of May 6th. Alright. So let’s start with the spicy stuff y’all. So OpenAI is rumored to launch a new search engine based on chat gpt technology. OpenAI is speculated to be launching a new search engine based on chat gpt technology, potentially challenging Google’s dominance in the search in in the search market and, maybe kind of taking this perplexity, approach. So the new search engine is rumored to focus on natural language interactions similar to perplexity, and Google’s search generative experience aiming to provide live data through AI.

Jordan Wilson [00:02:37]:
So reports suggest that OpenAI may may make a significant announcement on May 9th regarding this new search engine hinting at a potential game changer in the search engines, in the search industry. So OpenAI CEO, Sam Altman, said that he believes that integrating search into chat gbt could revolutionize the way that people find and act on information, offering a more efficient alternative to to traditional search engines like Google. So the intersection of large language models and search remains unexplored territory, I’d say, for the most part with potential for a more seamless and effective search experience for users. So, this could be a strategic move actually for Microsoft, right, which, reportedly owns 49% of OpenAI and is a major investor, in OpenAI. And, they’ve failed to challenge Google for search dominance. Right? So, it seems like no matter what, you know, Microsoft has done with its Bing search engine over the years, they’ve barely been able to crack any of the share. Right? So, depending on what study you look at, Google has historically owned about 90 to 95%, of the US search market. So this, I think, is a pretty big play that people aren’t necessarily paying attention to, especially with, you know, an analyst and a lot of enterprise users being a little underwhelmed by Google’s AI offerings, including their search generative experience, which, by the way, I personally like Google’s, SGE.

Jordan Wilson [00:04:10]:
Not a huge fan of their Gemini, you know, large language model. For whatever reason, Gemini has a problem talking to the Internet even though it’s powered, by Google. But, you know, this one’s pretty something else that’s worth noting is the timing of this. Right? So May 9th is is just rumors. Right? Nothing has been confirmed. There is technically no announcement, but I think this is literally going to be the, AI story of the week, you know, barring anything else that comes out if this does happen. Right? Not only could this be a major threat to Google, a huge, you know, bolster to Microsoft, but it could also ultimately change the way that we all use search engines. Right? If you’ve listened to the everyday AI show more than once, you’ve probably heard me talk about and other guests talk about how unusable the Internet is, you know, now and and how bad search engines are.

Jordan Wilson [00:05:05]:
Right? And how actually, how much of a time waster traditional searches. Right? And it’s kind of built that way. Right? Search engines are built on display ads. They’re built on taking you down rabbit holes. Search engines want you to click on 3, 4, 10 pages to get a result because they make more money and so do the online publishers. So, you know, as we saw, perplexity, come into the the the fold, you know, a little more than a year ago and essentially do that work for you. Right? So if you haven’t used perplexity, if you’re listening to the show, I highly advise you go do that now. Like, go ahead, click pause, you know, come back later.

Jordan Wilson [00:05:42]:
It doesn’t matter to me. But essentially, you know, it’ll look through 10, 15, 20 sources and try to do all that research for you. So this is kind of the area, where OpenAI and its CEO, Sam Altman, say that they want to play. Also, the timing is is very, I won’t say suspicious here, but it’s probably very intentional. Again, this is just a rumor, but, Google does have their big IO developer conference next week on May 14th. So that’s kinda why, you know, this this May 9th date that was, you know, first leaked out by a very famous, Twitter, tipster who has a highly successful track record, you know, supposedly out of Silicon Valley, you know, saying this May 9th date. But, OpenAI CEO, Sam Altman has already alluded to how OpenAI may want to, have its place in the future of search in a conversation with with Lex Fridman, last month and talked about cracking the code on search, but not in the way you might think. So he essentially Sam Altman said that Google search is just showing people 10 blue links and that there might be a better way to find that information.

Jordan Wilson [00:06:52]:
And he did say that OpenAI is very interested in helping people find better ways and more useful ways to find that information. So he he did kind of, you know, allude to that and said that he wanted, to kind of be a part of that, solution. So it should be pretty interesting. And a couple things to note, you know, are these early bread crumbs? Maybe. But OpenAI, did completely, change their website design last week. And something people aren’t really talking about is its focus on what looks like search. Right? So, if you’re if you’re on the live stream here, you you know, I’m showing you a screenshot of OpenAI’s website, but it looks like Google. Right? Weird.

Jordan Wilson [00:07:38]:
So, you know, OpenAI’s website previously was, you know, it just had a lot of information on it. It looked like a standard, it looked like a standard landing page. Right? It looked just like a tech company putting out information. So now if you go to open AI’s website or chat GPT’s homepage, you get redirected to the same one. And essentially, it’s just a box that says, ask chat GPT anything. Right? So it looks very much like Google. Right? A very simplified, a very simplified homepage. It wants you to go in and search and to talk to it conversationally.

Jordan Wilson [00:08:12]:
So a couple, a couple other key indicators, pointing to why this is likely going to happen or, you know, whether it’s May 9th or next month or, you know, next quarter. I I do think that this is gonna happen sooner rather than later. Right? So not only now do you have their home page, which is one of the most, visited websites in the world pushing people just to use chat gpt, and just OpenAI, in a conversational way. But, you know, now you don’t even need an account to use chat gpt. That’s another thing that I think is is definitely a nod to the direction that OpenAI is going with this. Right? Just like you don’t need, an account, to use Google. Right? Go to google.com, start searching. Right? So if you go to openai.com, you can start using chat g p t.

Jordan Wilson [00:09:00]:
Right? It’s the free version. So it’s not that great. It’s not going to, it’s really not good at actually searching the Internet in a traditional way. But, you know, I I I do think that this is, you know, this new homepage design that is really geared toward looking like a search engine will lead, to this. Right? So again, I I I don’t like to talk about, you know, projections a lot on the show. I like to keep it all news based, but, you know, this is where all the signs points, you know. Also, you know, some some tipsters online find, found this and, you know, we’re constantly monitoring this site now, but, they did also register a new kind of subdomain. So search.chatgpt.com, you know, and a lot of people on the back end have been looking at chat gpt and noticing this as well.

Jordan Wilson [00:09:47]:
So, yeah, this is definitely something you have to keep an eye on. Another thing, right, is there was this mysterious new large language model, that was out in the wild on the chatbot’s, arena that we talked about last week. Hey. That’s why you gotta, like, listen and watch every single day y’all because we cover this so you don’t have to spend hours doing it. But there was a mysterious model that was, you know, reportedly from OpenAI. It was kind of available for people to quote unquote play with and to test for about 2 days and then it was taken down, off this, the the chat board arena. And ultimately, what a lot of people believe this was is it was called the GPT 2 dash chatbot. And a lot of people said, oh, this is the next version of, you know, you know, GPT 4.5 or it’s GPT 5.

Jordan Wilson [00:10:35]:
And I said, no, I don’t think so. You know what? I ultimately thought it was one of 2 things, but one of them might lend it to being used for a potential search engine. So it did appear to be a much smaller model, not a multi, you know, trillion parameter model. And why that’s important is if you start offering a service like this, you know, if if you you, you know, like a perplexity. Right? Or a chat g p t. Right? If you start offering a service like this, you have to have a new more powerful model, that takes up less compute. Right? In a recent trend, we’ve seen this with Meta’s, Llama, with with Microsoft’s PHY 3. Right? You’ve seen these very recent over the last couple of months, these recent models that are much smaller and very powerful and focus on high quality data, and just human fine tuning.

Jordan Wilson [00:11:26]:
Right? Which is important because if ChatGPT were to offer a a service like this and they were to essentially really go this Google route, Right? Presumably, more and more people are going to be using it, and they have to have a way, to kind of convert people from free users to signing up. Right? So so people who aren’t even signed up to go in to chat gbt, use it like Google. Oh, I like this. So at least I wanna sign up for a free account so I can save my searches. Right? Because if you don’t have a free account, you can’t save your searches. So so that piece is big. I think that this new GPT 2 chatbot model may be the new free version, of chat GPT. So, you know, maybe they’re gonna phase out this, 3.5, which isn’t very good, and maybe use this new GPT 2, chatbot version, which has a knowledge cutoff of November 2023.

Jordan Wilson [00:12:15]:
So integrating that with browse with Bing, I think that’s what we we could be looking at. Again, it’s conjecture right now, but it looks like that’s where all the signs are pointing. Alright. Don’t wanna spend too much time, on that. So let’s say we got we have actually so much Microsoft and OpenAI news this week, but, let’s now talk a little bit about these newspapers. Right? Why are these newspapers, suing OpenAI and others? Well, let’s talk about it. So major newspapers in the US are suing OpenAI and Microsoft. So, this lawsuit has been filed against OpenAI and Microsoft and then for alleged copyright infringement.

Jordan Wilson [00:13:01]:
So a group of 8 US newspapers, including the New York Daily News and the Chicago Tribune is suing OpenAI and Microsoft for using copyrighted news articles without permission or payment to train their AI chatbots. So the newspapers claim that OpenAI, Microsoft have been, using, quote, millions of their articles to build their businesses without compensating them. OpenAI has stated that they are engaged in partnerships with news organizations to address concerns and to provide solutions. The lawsuit is a part of series of copyright lawsuits against OpenAI and Microsoft, with authors like John Grisham and media outlets also filing similar claims. Tech companies are essentially arguing, though, that using publicly accessible Internet content to train AI systems falls under the fair use doctrine of American copyright law even though the content is copyrighted. So OpenAI, it’s worth noting in the past couple of months has, struck up a lot of huge licensing deals with various media companies, including the Associated Press, Al Axel Springer, Priza Media, Lamonde, and the Financial Times among others. But, these lawsuits highlight the ongoing debates surrounding the use of copyrighted content by tech companies for AI training and its implications in the media industry’s intellectual property rights and the fair use of digital content. Also, it brings attention to the need for clear guidelines and arguments between content creators and AI developers to ensure fair compensation and ethical practices in the evolving landscape of artificial intelligence technology.

Jordan Wilson [00:14:40]:
So so more or less if you’re, you know, let me just super simplify it for you. But for the most part, large language models just scrape the entire Internet and use all of that data to train their models. Right? And guess what? And I won’t say an overwhelming majority, but a lot of that content is obviously copyrighted. So, you know, whether we think that this is right or wrong, it doesn’t matter. But, you know, copyright law is kind of, it’s kind of on the, on the title card. Right? Like this is going to be, being debated in courtrooms throughout the United States and throughout the rest of the world to say, hey, what does copyright law even mean anymore? Right? If it’s out there on the Internet, regardless of if it’s, copyrighted material, a lot of these, you know, companies such as OpenAI are saying, nope. We’re gonna go ahead. We’re gonna train, we’re gonna train our model on all of this information, and we’re gonna go down this fair use doctrine.

Jordan Wilson [00:15:35]:
And in the meantime, also try, you know, strike up huge, you know, multimillion dollar partnerships with content agencies. So this does become less of a worry in the future. Also what’s really important to to talk about and understand here is the New York Times case. Okay. So this is extremely important. So the New York Times sued, OpenAI and Microsoft back in December 2023. It’s still ongoing. So the, OpenAI tried to get the lawsuit dismissed a couple months ago, didn’t happen.

Jordan Wilson [00:16:08]:
But, the New York Times is alleging that OpenAI and Microsoft infringed on their copyright and used 1,000,000 of their articles, without permission, right, to train their models. And, the New York Times is seeking an unspecified amounts, but we’re talking 1,000,000,000 of dollars in damages. So, yeah, 1,000,000,000 with a b. So this one’s interesting and, you know, I’ve I’ve talked about this here on the show multiple times before. Kind of this this is the first big domino to fall. Right? This New York Times versus OpenAI lawsuit. I actually think it is going to impact, how we consume information for the foreseeable future. Right? A lot more than we may think.

Jordan Wilson [00:16:53]:
You know? In theory, this could go to trial. I don’t think it will actually go to trial. I don’t see how it is in any anyone’s best interest for this, you know, for this to go to trial. I do think that ultimately, this will be settled, out of court with a settlement. But I don’t see if I’m being honest, I don’t see how OpenAI and Microsoft win this lawsuit. I do think the New York Times made so many so many mistakes, which, you know, had a whole hour long podcast going over piece by piece, bit by bit, line by line of the New York Times case, which was, I think, shoddy at best. They did not do a very good job presenting their evidence in the court case, but that’s beside the point. But I really don’t see any other way, that this ends except for, a settlement being reached.

Jordan Wilson [00:17:42]:
And then what does happen? Right? So, now you’ve kind of seen this second wave of newspapers, some pretty pretty big names, in the fold there with the, the New York Post and the Chicago Tribune. So, you know, those are 2 of the top ten newspapers in the US kind of falling in line after this New York Times. So, I’ve talked about this on the show many times, but expect to see a ton of lawsuits, especially, once we get word of what is happening with the New York, New York Times OpenAI. And and maybe, who knows, but y’all, the news industry is extremely tight, extremely close. So when I did see this story last week and we talked about on the show, I immediately thought, okay, does someone there in the know, you know, know that something is coming down the pipeline with this big New York Times versus OpenAI and Microsoft lawsuit? Maybe. Right? There’s people who find, you know, leaks or who can, you know, get scoops on, you know, information and where stories are heading. It’s journalists. Right? Like, the former journalist in me says that, but that’s kind of how the world works.

Jordan Wilson [00:18:46]:
Right? Journalists and for the most part are always the people breaking the story. So you also have to think or wonder to say, okay. Does this mean that, someone at one of those, you know, 8 major newspapers knows. Right? Do they have connections at the New York Times, at the New York Times publishing company, knowing which way that this lawsuit is ultimately going? Because these things could take months, quarters, or years, to to finally, settle or if anything does actually go to trial. So pretty interesting there on the timing of everything. I honestly did not expect any other big name publications, to kind of launch their lawsuits, but they’ll all be launching lawsuits. Right? Literally, like, if a company out there does not yet have, a content partnership, with an OpenAI, with a Google, with an anthropic, etcetera, they’re probably going to be suing them. Right? So, we’re going to probably see, dozens or hundreds of these big name lawsuits from, newspapers, news organizations, content publishers, etcetera, in the coming months.

Jordan Wilson [00:19:52]:
But it was kinda interesting to me that, this one came before there was any big news on the OpenAI versus New York Times. So you have to think that there might be something in the coming, weeks or months coming out of that case, which which should be pretty interesting. Alright. Our next piece of AI news, which, hey, I’m human. I actually think I forgot to put the slide in here. So for our livestream audience, just follow along. But how did the Microsoft and OpenAI partnership actually happen? Was it was part of it a fear of Google? Well, maybe. So, a new European Union investigation has shown that Microsoft’s involvement in OpenAI might have been from a little bit fear of Google.

Jordan Wilson [00:20:38]:
Alright. So the European Union is currently investigating Microsoft’s investment in OpenAI, and that has prompted, the release of emails from Microsoft highlighting Microsoft’s concerns about Google’s AI dominance. So an email from Microsoft CTO, Kevin Scott emphasized the urgency from Microsoft to advance in AI, particularly in natural language processing to keep up with Google’s advancements. So the email at the time was sent to Microsoft CEO, Sadia Nadella, and Microsoft cofounder Bill Gates, highlighting Microsoft’s lag in AI compared to OpenAI in Google. So this email was from 2019. So in the email, Scott acknowledged, that maybe Microsoft underestimated the infrastructure investments made by Google and OpenAI in their AI developments. So this, reportedly, according to these emails, was a a big key in Microsoft’s eventual substantial investments in OpenAI, reportedly totaling between $11,000,000,000 $13,000,000,000 and that has really propelled development in this generative AI technology, notably the widely used chat GPT chatbot and Microsoft’s copilot, offering now which is bringing generative AI to the desktop, which is largely based on OpenAI’s GPT 4 model. Right? So this is pretty interesting when this story came out.

Jordan Wilson [00:22:05]:
You know, essentially, Microsoft saying about 5 years ago, like, hey, we’re behind. Look at what Google is doing. Look at what OpenAI is doing. We are not doing enough. So, you know, obviously, this has led to the integration inside of Microsoft’s products for a lot of, OpenAI and and gpt 4, led products including, you know, their browse with Bing, GitHub, Microsoft 365 products, etcetera. So so it’s it’s gonna be interesting to follow, this kind of investigation and what ultimately becomes, of this European Union, investigation into Microsoft’s investment in OpenAI. You know, because the the outcome holds significance beyond just the involved companies potentially reshaping the trajectory of AI research and development on a global scale. What’s also interesting if you look at just that that, you know, blip in time on the radar in 2019 and then you fast forward to today and you have to wonder if that email was never sent.

Jordan Wilson [00:23:08]:
Right? If Microsoft never would have made this substantial investment, reportedly giving them a 49% ownership stake in OpenAI. So you have to think, well, what what would what would have happened if that never would have happened. Right? But right now, if you’re a Microsoft shareholder, you’re probably pretty thankful that it has happened, because a lot of analysts and smart people have said that, you know, a lot of what Microsoft has been able to accomplish over the last year or 2. And a lot of their success has come from their investment into generative AI technologies and a pretty successful rollout of their copilot offerings across their Microsoft 365 suite of products. Right? And also interesting, since 2019, Microsoft’s stock is up more than 220% since those emails were revealed, and that is far outpacing the growth of, you know, Google and their parents company, Alphabet stock during the same time. So, yes, Google is growing. It’s still one of the largest companies in the world. But, hey, right now I would say Google is chasing Microsoft.

Jordan Wilson [00:24:12]:
Right? Which, you know, we just talked about Google’s search dominance, but you probably wouldn’t have thought maybe 5 years ago that Microsoft would be in the position that they are today and maybe a lot of it does stem from some of those emails. Alright. Let’s keep this going. So this is some some new news. Right? Sometimes we we kind of recap the news for the week, but the Biden administration has just allocated $285,000,000 for semiconductor digital twin development. Alright. So the Biden administration is set to fund efforts enhancing semiconductor manufacturing through digital twins, virtual models for testing and optimizing physical objects in systems. So the funding totals $285,000,000 will support research into semiconductor, digital twin development, combined physical and digital facilities, industry demonstration projects, workforce training, and the establishment of the CHIPS Manufacturing US Institute.

Jordan Wilson [00:25:09]:
Alright. So advancements in semiconductor digital twin technology supported by this funding can potentially lead to a lot of AI driven innovations in chip development, manufacturing processes, and collaborative design efforts within the semiconductor industry. The funding is part of the CHIPS and Science Act of 2022, which is a $280,000,000,000 bill aimed at boosting domestic semiconductor manufacturing. So, you can’t really talk about this huge piece of news. 285 $1,000,000 essentially for digital twins without quickly talking about what are digital twins and how are they being used right now in AI. Well, it’s it’s hard to talk about digital twins without talking about NVIDIA. Right? And when you talk about semiconductors, you can’t not talk about NVIDIA. Right? So essentially, if we’re gonna oversimplify it here y’all, but, generative AI needs a lot of computes.

Jordan Wilson [00:26:05]:
It needs a lot of computer power. Right? And a lot of these are highly, specialized and highly specific, GPU chips that NVIDIA is leading in. Right? So the rest of the semiconductor industry has been scrambling over the last couple of years as NVIDIA has gone from a relatively unknown, you know, video game graphics company to one of the biggest companies and one of the most important companies in the world. Right? Depending on which day you look at it, they’re normally a top 3 to top 4, company by market cap here in the United States, right, after not even being, you know, a top 30 company about 4 years ago. So this does, you know, this move from the Biden White House does signal, kind of the increased need for other companies to stay competitive as well. Because right now, a lot of the other semiconductor, you know, companies here in the US and otherwise are struggling to keep up with NVIDIA. So I I I do think that this is this move partially is to bolster competitiveness here in the US, which will ultimately and hopefully drive down prices. Right? Because right now, you know, if you’re a big generative AI company or if you’re just an enterprise company really wanting to, incorporate generative AI into your offerings, you gotta have a lot of money.

Jordan Wilson [00:27:22]:
Right? A single, you know, Blackwell GPU is not gonna do anything. You’re probably looking at a 7 to 8 figure investment if you really wanna get generative AI, you know, going at an enterprise company. So this this move is pretty big and also just talking about digital twins. Right? If you don’t know digital twins, those aren’t just avatars, you know, of yourself, and it’s like, oh, that’s an AI talking. No. Digital twin essentially think of it as a city, or a manufacturing plant. Right? So that’s before you spend, you know, 100 of 1,000,000 of dollars or maybe 1,000,000,000 of dollars building, you know, a giant new facility, you know, huge pieces of infrastructure. Essentially, a digital twin environment allows you again, you have to use a lot of, you know, generative AI powered by these very powerful GPUs, but it essentially allows you to simulate everything millions and millions of times in a digital twin environment.

Jordan Wilson [00:28:15]:
So think of it literally like a city. Right? I live in Chicago, so in theory, a city could come in and create a digital twin of Chicago and they could run simulations millions of times a day, and find different ways to improve energy, find different ways to improve, you know, highways, you know, find different ways to improve infrastructure, etcetera. Right? But you need a lot of compute power. So, this this, kind of news from the Biden administration might not sound like a lot because it’s like, alright. Well, whatever this semiconductor digital twin development is. But for the US government to be investing nearly $300,000,000 into essentially digital twins, is a pretty big development. Alright. Our next piece of AI news that matters for the week, as I take a sip of water.

Jordan Wilson [00:29:05]:
Warren Buffett has warned of AI threats at the annual shareholders conference for Berkshire Hathaway. So famed investor Warren Buffett during the annual Berkshire Hathaway shareholders conference compared the rise of AI to the dangers of nuclear weapons, expressing concerns about its potential impact. Pretty pretty big shift there, Warren, but we’ll talk about that here in a second. So Buffett highlighted the significance of AI acknowledging its importance and the potential for both good and harm. So the Berkshire CEO shared his unease after encountering a video generated by AI using his likeness. Yeah. We are just talking about digital twins, but there we’re talking deep fakes. So, you know, after encountering a video generated by AI using his likeness to deliver a message, he never conveyed racing fears of potential scams in the future.

Jordan Wilson [00:29:57]:
So despite recognizing AI’s vast potential, Buffett admitted his lack of understanding about the technology and its unpredictable outcomes. So this highly anticipated conference in Omaha, Nebraska is known as the Woodstock for Capitalists, and it featured Buffett alongside a couple other vice chairs where they discussed Berkshire’s financial standing and future prospects. Here’s here’s a hot take. Well, I’m I’m glad that, you know, finally, Buffett admitted his lack of understanding about the technology and its unpredictable outcomes because previously, in previous years, Warren Buffett has really underplayed what generative AI and what AI is actually capable of. So, you know, I think, it was more than more than a year ago, he, you know, essentially said that this large language model and generative AI technology was extraordinary, but wasn’t sure if it was beneficial. And the beneficial is in quote, right, which is asinine to me to think that someone as smart as Warren Buffett, even a a year ago, could look at large language models and generative AI technology. And even at that time, right, we’ve talked about it here on the show as well, the magnificent 7, especially minus Tesla. So, you know, 6 of the largest companies in the US have been pulling the weight of the entire US economy at, you know, ways we’ve literally have never seen before in the history of the world.

Jordan Wilson [00:31:19]:
Right? So, the fact that Warren Buffett, it took him this long to finally realize, oh, okay. Yeah. This AI technology actually might be pretty good and it might be so good. It’s kinda scary. Right? Yeah. Because, you know, a year ago, he was singing a completely different tune, kind of talking about AI that it was, you know, just a toy. You know? He said, ah, it’s extraordinary, but not sure if it’s beneficial. Yeah.

Jordan Wilson [00:31:39]:
It’s beneficial, and it’s changing the way we all work. So it’s good to see that, you know, Warren Buffett may finally be on board, but it just took a deep fake of him, to to realize the just how, quote, unquote, beneficial, generative AI technology is. Alright. Our next piece of AI news that matters is Eli Lilly scientists have just said that generative AI will be help discovering drugs. Alright. So scientists at Eli Lilly just recently said that the use of generative AI let, led to the discovery of novel drug candidates with, quote, weird looking structures that traditional methods missed. So if you don’t know, Eli Lilly is a pharmaceutical giant that specializes in developing and producing medications to treat various health conditions. So this article just came out, but said, scientists were surprised by the AI generated decisions sparking creativity and opening new pathways in medicine development.

Jordan Wilson [00:32:39]:
So the field of AI and health care is moving toward completely AI generated medicines in the near future, which is wild to think about. Right? And when people say, oh, you know, generative AI technology, you know, can’t, you know, create new intelligence. Well, this just goes to show you, yeah, it can. You know, this is one of the largest pharmaceutical companies, pharmaceutical companies in the world saying that large language models and generative AI will ultimately lead to discovering new medicines that human scientists have not been able to discover. Also, I mean, you can’t talk about this without talking about Google, Google’s DeepMind’s AlphaFold breakthrough in 2021 as that marked a pivotal moment in applying AI to biology, revolutionizing drug developments and design. What’s also interesting here is, Eli Lilly’s scientist talked about hallucinations and how hallucinations in this case are actually a good thing. And that hallucination so when a large language model maybe doesn’t quite understand and it kind of goes off the rails, they’re talking about, hey. Hallucinations may actually, in large language models, may actually help us, discover or it might lead to help us lead to the discovery of new medicines, or new breakthroughs in medical technology that we didn’t know before, that we couldn’t see before.

Jordan Wilson [00:33:56]:
So they did talk about AI’s ability to hallucinate new drugs by exploring a vast number of proteins could loot could lead to groundbreaking discoveries. Right? So we talk about hallucinations like they’re a bad thing and, you know, for the most part, they are unless you are using hallucinations intentionally. So we’ve talked about that here on the show before. But I do think that that is a pretty important piece. Right? And how, yes, generative AI and large language models, are technology that can create new intelligence because when you can, run just like we talked about running, you know, millions of simulations in a digital twin environment, this is what, medical companies are doing with protein discovery, with with medicine. Right? When you can run at a very high level and as these, generative AI models get smarter and smarter and more capable, when you can run millions of simulations in literally, a fraction of the time that it would take, even if you brought together all of the world’s best and brightest scientific minds, you can accomplish so much more with generative AI models. Right? Like, we’ve had, you know, large, you know, large tech, and and pharmaceutical leaders in the world come on this very show and say as much. So, I I do think, you know, even you like, you might look at this and kind of laugh it off.

Jordan Wilson [00:35:13]:
Like, hallucinations aren’t gonna lead to new drugs. Well, yeah, they are. They’re they’re definitely gonna lead to news drugs. Generative AI and large language bottles and the ability to essentially have an unlimited, amount of of resources of, you know, all of the top knowledge in the world. And and to be able to, you you know, run that at scale is extremely exciting for the future of medicine. Alright. That’s not it. Well, that’s it for today.

Jordan Wilson [00:35:42]:
Hey. Like, love what Cecilia said here and thanks for our livestream audience for joining, saying hallucinations often equal, extra pollutions which generate innovations. Man, I can’t I can’t speak today y’all. But I hey. I wanna know what you think of of this AI news. So if you are on the podcast, we always put in, an email, like my email, my, LinkedIn. Let me know what do you think of this AI news. And just just to recap, because there was a lot.

Jordan Wilson [00:36:14]:
So here’s the AI news that matters. We’re looking at OpenAI and ChatJPG potentially, releasing a search engine maybe as soon as this week. We have some new information from this European Union that’s investigating Microsoft’s investment in OpenAI and some emails kind of saying it was their original fear of Google. The Biden administration allocating $285,000,000 for semiconductor and digital twin developments. 8 more, major US newspapers suing, ChattGPT’s OpenAI and Microsoft. Warren Buffett finally saying, oh, yes. This AI is actually very capable, and it might be a threat. So talking about that at the annual Berkshire Hathaway conference and researchers from Eli Lilly recognizing the extraordinary power of generative AI when it comes to helping them, find, develop, and discover new medicines.

Jordan Wilson [00:37:12]:
Alright. I hope this was helpful. If it was, hey, leave me a comment. If you could, share this with your friends as well. Right? We spend hours literally every single day going through what’s happening in the world of AI, bringing on experts, you know, putting out tutorials to help you grow your company and to grow your career. So if this is helpful, it takes about 5 seconds to, you know, tag a friend if you’re on, LinkedIn. Click that repost button. Or if you’re listening on the podcast, please leave us a review rating or tell your friends.

Jordan Wilson [00:37:40]:
Alright. Thank you for tuning in. We hope to see you back tomorrow and every day for more everyday AI. Thanks, y’all.

AI [00:37:50]:
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 sign up to our daily newsletter so you don’t get left behind. Go break some barriers, and we’ll see you next time.