Everyday AI Podcast – An AI and ChatGPT Podcast - EP 576: GPT-5 release timeline update, Google and Microsoft vibe coding and more AI News That Matters

Episode Date: July 28, 2025

Could GPT-5 only be weeks away?Why are Microsoft and Google going all in on vibe coding?What's the White House AI Action Plan actually mean?Don't spend hours a day trying to figure out what AI means for your company or career. That's our job. So join us on Mondays as we bring you the AI News That Matters. No fluff. Just what you need to ACTUALLY pay attention to in the business side of AI. Newsletter: Sign up for our free daily newsletterMore on this Episode: Episode PageJoin the discussion: Thoughts on this? Join the convo and connect with other AI leaders on LinkedIn.Upcoming Episodes: Check out the upcoming Everyday AI Livestream lineupWebsite: YourEverydayAI.comEmail The Show: [e-mail] with Jordan on LinkedInTopics Covered in This Episode:GPT-5 Release Timeline and FeaturesGoogle Opal AI Vibe Coding ToolNvidia B200 AI Chip Black Market ChinaTrump White House AI Action Plan DetailsMicrosoft GitHub Spark AI Coding LaunchGoogle’s AI News Licensing NegotiationsMicrosoft Copilot Visual Avatar (“Clippy” AI)Netflix Uses Generative AI for Visual EffectsOpenAI Warns of AI-Driven Fraud CrisisNew Google, Claude, and Runway AI Feature UpdatesTimestamps:00:00 "OpenAI's GPT-5 Release Announced"04:57 OpenAI Faces Pressure from Gemini07:13 EU AI Act vs. US AI Priorities12:12 Black Market Thrives for Nvidia Chips13:46 US AI Action Plan Unveiled19:34 Microsoft's GitHub Spark Unveiled21:17 Google vs. Microsoft: AI Showdown25:28 Google's New AI Partnership Strategy29:23 Microsoft's Animated AI Assistant Revival33:52 Generative AI in Film Industry38:55 AI Race & Imminent Fraud Crisis40:15 AI Threats and Future InnovationsKeywords:GPT 5 release date, OpenAI, GPT-4, GPT-4O, advanced reasoning abilities, artificial general intelligence, AGI, O3 reasoning, GPT-5 Mini, GPT-5 Nano, API access, Microsoft Copilot, model selector, LM arena, Gemini 2.5 Pro, Google Vibe Coding, Opal, no-code AI, low-code app maker, Google Labs, AI-powered web apps, app development, visual workflow editor, generative AI, AI app creation, Anthropic Claude Sonet 4, GitHub Copilot Spark, Microsoft GitHub, Copilot Pro Plus, AI coding tools, AI search, Perplexity, news licensing deals, Google AI Overview, AI summaries, click-through rate, organic search traffic, Associated Press, Condé Send Everyday AI and Jordan a text message. (We can't reply back unless you leave contact info) Start Here ▶️Not sure where to start when it comes to AI? Start with our Start Here Series. You can listen to the first drop -- Episode 691 -- or get free access to our Inner Cricle community and all episodes: StartHereSeries.com Also, here's a link to the entire series on a Spotify playlist. 

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. Meet Firefly AI Assistant, now live in Adobe Firefly, the All In One Creative AI Studio. Just describe what you want to create and the assistant handles the rest, orchestrating multi-step workflows across Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome. The assistant accelerates execution. We have our first official reporting on a potential GPT-5 release date.

Google and Microsoft are going all in on vibe coding. And the White House AI action plan is as controversial as it is bold. You can spend hours every single day trying to keep up with what's happening in the world of AI and how it impacts your company or career or wild idea here. Just spend your Mondays with us as we cut through the fluff, the marketing, and the hype, and bring you the AI news that matters. All right. I'm excited for today's episode, a ton going on in the world of AI.

Hope you're ready for a quick hitting recap too. What's going on, y'all? If you're new here, welcome. My name's Jordan Wilson, and this is Everyday AI. This is your daily live stream podcast and free daily news. that are helping everyday business leaders like you and me, not just keep up with AI, but how we can leverage this all to get ahead

to grow our companies and our career. So if you're trying to be the smartest person in your department when it comes to AI, it starts here, the unedited, unscripted podcast and live stream, but where you take it to the next level is at our website, your everyday AI.com. So there, not only is it like a free generative AI university with hundreds of leading experts in the world,

you can go listen to whatever you want. It's all on our website, all for free. But also make sure you go sign up for our free daily newsletter where we recap not just everything happening in the AI news, but each day's podcast episode. It's all in there. So yeah, we do the AI news on Mondays, most Mondays, hot take Tuesdays, putting AI to work Wednesdays and usually some interviews on Thursdays and Friday. So I'm excited. Let's get into without further ado, the news for July 28th this week. And hey, what's up?

Live stream audience. It's good to see you. We'd love to hear your thoughts and I'll share them live as we go along. So our friends joining us on YouTube, big bogey face and Nathan, good morning to you all. And Frank and Dan and Ted, everyone else, thanks for joining us, people on the LinkedIn machine. Dr. Scott, Marie, Joe, everyone else, thanks for joining us. We'd love to hear your comments as we go along. So our first piece of AI news, pretty big one here.

We have our first official. reported date anyways for the highly anticipated GPT5 model from Open AI. So according to the verge, Open AI is expected to release its next generation GPT5 language model in early August. So this release is very notable because it arrives ahead of previous expectations and before the day of what we've been hearing is a potential Open AI open AI open. language model. So open AI CEO Sam Altman confirmed on Axe or Twitter, as most people call it, that GPT5 will be released soon and showcase the model's advanced reasoning abilities in a recent podcast appearance. So according to reports, Altman described GBT5 using it as a here it is moment

in terms of AGI. When GPT5 instantly answered a question, he could not, based on his own personal context highlighting the model's leap in problem solving. So GPD5 will reportedly include a mini and a nano version, which will be accessible also via OpenAI's API, broadening its reach for developers and businesses. So the new model will integrate OpenAIs, O3 reasoning capabilities, which were originally planned for a separate release. So they're kind of stream, trying to anyways streamline the company's product lineup. Also, we've seen kind of reports in Microsoft co-pilot of a forthcoming GPT5 model selector inside of the web version of Microsoft co-pilot as engineers there have reportedly been preparing server capacity for GPT5 since earlier this

year. Also, there's been some models that have popped up on different model evaluation platforms, such as the LM arena, where you can go and test out different upcoming models. Some of them, named, some of them codenamed. So a lot of reported code named models have been popping up and sometimes getting deleted, including ones named Summit, Zenith, Lobster, Nectarine, and Starfish. I was able to play with these just a little bit. Sometimes it's a roll of the dice, quite literally, to be able to get these random model selectors to be going in a head-to-head battle. That's the only way that they're going to come up while using L.M. Arena. And very impressive results. so far. So if those truly are GPT5 or variants, right, whether you have a full or a

nano or mini version, pretty impressive so far. And this has been one of those things. The GPT4 technology is now more than two years old. So Open AI, and I've been saying this now for like 18 months, they've had no real reason, I think, to release GPD5, at least over the past year or so. I'd say that's probably changed because it's always been one of these things where open AI has been able to essentially release an updated version of its GPT40 model. Again, based on the two plus year old GPT4 architecture. And essentially when they have recently updated their GPT40 model on the hood, it's benchmarked very well, usually a top one or top two model. And I think that's changed, at least with their base model, because the Google variants of Gemini 2.5 Pro have been dominating over the last two to three months. So maybe Open AI is finally starting to feel the pressure of not having what's widely considered the best all-purpose model with Gemini 2.5 pro being a reasoning model.

So we've heard reports now for more than a year that the GPT5 will be more of a system. and it's going to kind of select the best model. So it might select as an example, O3 Pro. It might have some new variants of GPT5 replacing GBT40, but we could now find out actually in a week or so, right? Here it is the last week of July, and we're seeing confirmed reports now, right? We've heard mumbles for the last year. Oh, GPT5 is coming out.

And I've been saying, no, no, it's not. I think this is the first time when I've seen a legitimate report from a news organization saying, yeah, like, yeah, this is probably something that could come out in the coming weeks. So pretty big news here. Dr. Scott asking, do you think the GPD5 release or other company releases will beat the EU AI Act 82 implementation, extending their time for compliance to 827? I'm not sure, but if I'm telling you to the truth, I don't think OpenAI, at least right now,

is that concerned about the EU AI Act because it's the U.S. right? That's where the enterprise business customers are. That's where the appetite is. That's where the growth is. Obviously, there's other countries that are, you know, they're concerned about in their long-term plans, but you want to have the best model for the U.S.

because that is where the majority of enterprise customers are located, and that's where the money is. It's in the enterprise customer. So, yeah, rollouts across different markets will be slower for sure. All right. Our next piece of AI news, vibe coding, Google going all in. So Google has started testing and has officially released Opel, a new AI powered tool that lets users build many web apps from a simple text prompt.

So Opel is available through Google Labs right now in the US and is designed to help anyone, even those with no coding experience, create and share web apps quickly. So users can describe the app that. they want in natural language and Opel uses Google's AI models to generate it, then allows further customization through a visual workflow editor. So the tool offers a gallery of remixable apps as well. So if you don't know how to get started, that's a great way to do it, making it pretty easy to start from existing templates.

And once an app is built, this is pretty cool, users can publish it online and share links for others to test it using their own Google accounts for access. So Google's, and that's a pretty big piece there. right so you can literally embed google's ai into an ai app that you create but then it's like okay well who's ultimately paying for it so as long as users have their own paid google gemini accounts they can use it so it's almost like you're spinning up a version of you know chat tvt or google gemini with specific features or functionality and you can use it instantly in production which is really cool uh so google's move here signals a push to make app development accessible

to a wider audience, not just professional developers. Also, this launch puts Google in direct competition with other no code and low code platforms such as cursor, lovable. Canva has their new AI app maker, Figma, Replit. I mean, there's dozens of other companies and startup tools that are getting into the vibe coding. And also, you know, Google has had other kind of vibe coding tools, right? But I think this is.

bringing it to an even less technical audience, if that makes sense. Also, I've been like pretty bullish on a simple feature that's been living inside of Google Gemini now for more than a month that we've shared about on our put AI to work on Wednesday's series. Right. So if you're using Google Gemini, there's actually a very little wand icon in the lower right hand corner. if you're using a paid Google Gemini account, you can essentially embed Google Gemini into any web app that you're building inside Google Gemini, super meta, right?

But it's been this little icon, and I've been saying for weeks now, that is one of my favorite features that I've seen of any platform and how I was kind of surprised that Google kind of snuck this out under the radar. And it seems like there were bigger plans for this all along.

So it seems like that little feature, or functionality inside Google Gemini is now getting a complete rollout via Opal, which makes sense, right? I'm like, this is too good to be true to be kind of a hidden feature inside Google Gemini. So we'll definitely, Opel, let me know, live stream audience, podcast people. You can actually leave a comment on the podcast. I can't reply to them. But let me know.

Opel, yes or Opel, no. Should we do a dedicated Opel show in the near future? I'll leave it up to you guys. unofficial vote via comments, right? So Opel, yes, Opel, no. Let me know if we should be covering it. All right. Our next piece of AI news.

According to the Financial Times, Nvidia's most powerful chips are still flooding China despite a US ban. So according to the financial times, at least $1 billion worth of band, Nvidia B200 and other advanced AI chips from the US-based chip maker. have been shipped to China since U.S. export restrictions tightened in April. So the U.S. government banned the export of China-specific H20 GPUs and similarly powerful AMD-M-I-308 chips. But Chinese companies have reportedly been able to continue to access these

products through secondary black markets and global trade loopholes. So specifically, if we're looking at NVIDIA's B-200 chip, which is one of the most powerful in the world and is especially in high demand globally, with prices in China's prices in China reaching up to 50% higher than US levels. So as an example, that's one rack of B200 GPUs is selling for $420,000 to up to a half million dollars netting those resellers, those people on the black market, over $100,000 in profit per sale. So a pretty lucrative kind of gray market or black market business in bringing these banned chips from the U.S. to China. So despite NVIDIA and its affiliates not servicing or supporting officially these chips in China,

growing demand and profit margins have fueled a thriving black market. So NVIDIA CEO, Jensen Wong has downplayed the scale of unauthorized chip sales, but he and former U.S. Commerce Secretary, Gina Romando, have both questioned the effectiveness of U.S. export controls of chip sales to China, warning that they may actually backfire by accelerating China's domestic AI hardware development. So U.S. authorities are pressuring allies to crack down on smuggling, but it looks like that this hasn't happened yet. So pretty, I won't say interesting because we all knew this was happening. But some worthwhile and helpful reporting that's coming out that's just showing the scale of this. Also, with the new White House AI action plan, this is extremely relevant.

Speaking of that, that is our next piece of AI news. So President Donald Trump's White House has unveiled a 90 plus action. So 90 plus different action items in their new AI action plan and has signed three executive orders on AI launching over 90 government actions to accelerate AI innovation and outpaced global competitors, especially in China. So this came in the middle of last week as the White House. officially unveiled their White House AI Action Plan, which is structured around three main pillars, which is speeding up America's AI innovation, building up their tech infrastructure, and strengthening U.S. leadership in international AI diplomacy and security. So I, well, it's pretty simple what they're doing here.

The three pillars are three eyes, innovation, infrastructure, and international leadership. So key actions in the White House AI Action Plan include updated. federal procurement rules to require agencies to purchase only ideological neutral AI. That's a, that's going to be a mess, removing federal regulatory barriers around AI and limiting federal funding to states that decide to regulate AI on their own. So one executive order entitled, Preventing Woke AI in the federal government, kind of or, you know, called that anyways, mandates that agencies only. use AI systems that are deemed unbiased by the administration and directs the Office of Management

and Budget to issue new guidance for AI assessment. The plan specifically targets the removal of references to misinformation DEI initiatives, which are diversity, equity, and inclusion, and climate change from federal AI guidelines with a senior official calling DEI the main target. The administration will consider a state AI regulatory climate when making funding decisions, which is definitely going to be challenged in court because this is going to reduce federal support for states that decide to legislate AI on their own. So previously, the Trump White House tried to pass this in their budget, which was nicknamed the big beautiful bill, but U.S. senators knocked that provision down by a 99.

nine to one vote. Right. So essentially they tried to pass this in legislation. It got shot down that said if states decide to make AI laws, they wouldn't be eligible for any federal funding. So now this is part of this executive order and the White House action plan that says the federal government can decide not to give a state federal funding. So yes, sorry, global listeners.

Things work a little differently here in the U.S., right? So different states, obviously. get funding from the federal government. So this allows the federal government to withhold funding if certain agencies are using what they decide is either woke AI or if individual states pass AI laws and regulations, right? So it gives the federal government sweeping authority, which we kind of broke this down. So I'm not going to get into all my hot takes on this one.

So if you want that, make sure to go listen to episode 574. from Thursday when we go over the AI action plan in more details. But yeah, expect actually a lot of Republican infighting on this one. So whatever Republican state decides to regulate and legislate AI first, there's going to be some federal funding fights and probably some legal challenges also because a lot of civil society organizations, including the Center for Democracy and Technology, have criticized the plan for overlooking responsible AI development

and for provisions that could penalize states and dictate what is considered truthful AI. So, yeah, I would assume a lot of legal challenges and a lot of Republican infighting on this one in the months to come. All right. Yeah, so a good question here. Adobe just introduced an entirely new way to create, bringing the power and precision of its creative. suite into one conversational experience. Meet Firefly AI Assistant, now live in the Adobe Firefly app, the All In One Creative

AI Studio. Powered by Adobe's Creative Agent, Firefly AI Assistant lets you start with your vision, just describe what you want, and shape the outcome as it takes form with the Assistant. The Assistant orchestrates multi-step workflows, drawing on 60-plus pro-grade tools across Adobe Creative Cloud apps, including Photoshop, Illustrator, Premier, Lightroom Express, and and more to help bring your ideas to life. You can also get started with creative skills, a growing library of pre-built workflows for common creative tasks,

like batch editing photos, creating mood boards, portrait retouching, and creating social variations. Every step the assistant takes is visible, so you can refine, redirect, or take over at any time.

You stay in the driver's seat as the creative director. Adobe Firefly AI assistant now in public beta. See it today at Fire. fly.adobby.com from Graham just saying like how do you think they're going to be able to police this? Yeah. It's going to be the wild west. Right. I think the White House, it depends on how much noise that they want to make and what else is happening on the political climate to tell you the truth. I have a little bit of a background in, you know, I covered politics, you know, for the Chicago sometimes at the federal state in a little, or sorry, at the state

level and a little bit on the federal side. So I do have a little bit of experience in this, but it honestly depends, right? It depends on how quickly some of these legal challenges are going to go through. If anything in here is ultimately going to get overturned, right? Because so much of what happened in the White House AI action plan was actually just by executive order and by essentially dangling federal funding in front of states. So we'll see how these legal challenges play out in the courts because there will be many legal challenges to the new White House AI Action Plan. All right. Next, Microsoft is following suit with Google getting in on the AI vibe coding. So Microsoft own GitHub has released a public preview of GitHub Spark,

a new AI powered feature that allows users to turn natural language ideas into full stack apps in minutes. So right now, Spark is available to get. GitHub co-pilot's pro plus subscribers for $39 per month or $390 per year, with users getting up to 375 messages per month and a limit of 10 active app building sessions at a time. So interestingly enough, this tool leverages Anthropics Claude Sonnet 4 as an available model and allows users, even non-technical users, to just use plain. English and also provide a visual editing environment or assist with traditional manual coding.

Once an app is published inside GitHub, users can publish it to or create a new GitHub repo with a single click, streamlining the process from idea to deployment. So interesting one here from Microsoft and GitHub, uh, making this only available right now to GitHub pro plus users, although, According to GitHub, they did say Spark will be made available to non-copilot Pro Plus users. That's a mouthful, by the way, in the coming weeks. So essentially, you have Microsoft and Google, the two biggest players in the enterprise tech space, going all in on the same week with vibe coding apps. Right.

I would be, if I'm being honest here, I'm going to give the edge to Google here, right? Because right now, even though I do think, obviously, more developers are using GitHub and using GitHub co-pilot right now. I do think Microsoft's pricing here in their rollout doesn't make a ton of sense, especially if you compare it head to head. So Microsoft's main advantage here is, well, GitHub, right? Literally, any developer uses GitHub. Right. So the GitHub repository and being able to work directly with that is a huge advantage to Microsoft co-pilot and GitHub as well. But I think Google with the pricing, making so many of their vibe coding tools available for free or free 99, at least for me personally, I'm going to be following their vibe coding tool a little more closely with Opel than I will.

GitHub co-pilot Spark. All right. More Google News. Our next AI news story is Google is now reaching out and trying to pilot a new licensing program amid search traffic decline. So Google is now reportedly reaching out to 20 national news outlets with licensing deals to train its AI models, according to reporting from Bloomberg, marking a significant change in strategy as competitors like OpenAI and

Plexity secure individual publishing partnerships. So this move comes as Google's share of the global search market has dropped below 90% in six of the past seven months, which hasn't happened in more than a decade. And that's according to similar web. So essentially, there's been a huge uptick in chat GPT search traffic, people using chat GBT and perplexity and other AI search engine or AI chat bots as search engines, which has cut into. Google's traditional search traffic. So now news publishers have expressed concern that Google's new AI overview features, which summarize news above traditional search results, could reduce their organic search traffic by up to 60% as users get information without clicking through to publisher sites.

And a new Pew Research study shows that traditional kind of blue links get about a 15% click rate, whereas those in AI. Summaries get less than that at about 7%. So these numbers with publishers saying they're losing about 60% of traffic are checking out on the research side as well. So according and with some examples, according to similar webs data, the Wall Street Journal's share of overall search traffic has fallen from 29% to 24% over the past years, while the New York Times share of organic search traffic in their individual websites has dropped

from 44% to 36%. And you might say, oh, that's a 5% percentage drop. Well, yes, it is, but those are roughly 20% drops individually in their own kind of makeup of traffic to their website, which is huge. So until now, Google's only major licensing deals were with the Associated Press and Reddit, while Open AI has partnered with a ton of publishers, including Kondaynast. dot dash the Atlantic and perplexity has deals with the LA Times and the independent as well as others. So Google's new pilot project, which includes initiatives like its pinpoint research tool,

journalism, AI innovation challenge and newsroom training for its Gemini app and other tools, signals a push for Google to repair relationships with publishers whose content is vital for Google, both search and for AI development. according to Google's blog post, the company is exploring and experimenting with new types of partnerships and product experiences, but has not shared specific details about its plan or conversations with publishers. And y'all, I've been saying this literally for two plus years. I said for major publishers, there's only three options. Number one, you strike a deal with Google, Open AI, perplexity, Claude, etc. That's number one.

number two, you sue them, right, which we've seen plenty. Or number three, you potentially go out of business. There's literally no other way that this shakes out. But this is the first time that we've seen a more partnership approach at large from Google with this new, this new kind of push going after 20 national news outlets at once. So we'll definitely be following this because, it's actually really important, not just for how we all search the web, because that's changing, and it's going to AI first, right? And we're using these AI chatbots and, you know, probably Google AI overviews or AI mode to do

business-related research, but then also on the publisher ed, right, so many companies, not even just online publishers, their models, right? Even look at everyday AI, for example. Our business model is we put out as much free information as possible. we hope that people discover it right on Google search or on YouTube whatever and then we hope that they subscribe to our newsletter right that's our way to monetize uh right however even our stuff we're right we're a smaller uh you know publishing company our stuff is in all the AI um you know search engines so it's it's something that affects even your business right you probably

don't know how your company right unless you're in marketing advertising or something like that But so many companies that even you wouldn't think, you know, thrive from SEO, thrive from lead gen, you know, from different marketing funnels. This is so many companies playbook. And it's really been flipped on its head right now because you're getting and your potential customers are getting all of these answers without ever visiting websites now. And we'll see how this plays out because, you know, perplexity and open AI, you know, as an example, are starting to roll out checkout features in app.

So you won't even have to leave, you know, Open AIs chat, GBT or perplexity to shop for things now. So, you know, there still will be a way, I think, for people who sell digital or sorry, who sell physical goods to still compete without sending users to their website. But this is actually a lot bigger than most people realize. All right. Our next piece of AI news, Microsoft's Clippy is kind of making an AI version come back because Microsoft has launched a very big.

visual avatar for its co-pilot AI chat bot, giving its co-pilot character or avatar, real-time, facial expressions, gestures, and conversational memory. So the freshly rolled out copilot's appearance is what it's called. Right now, it's available as an early preview inside of copilot labs and is currently limited to just the web version of co-pilot. So essentially there's a cutesy little a raindropped shaped little robot character that users can talk to. And you can activate the avatar by entering voice mode inside copilot online, clicking on the microphone icon, then titling on copilot appearance in voice settings. So when enabled, copilot responds to voice inputs with smiles, nods,

and other nonverbal cues, aiming to make conversations more natural and engaging. So Microsoft's AI team led by Mustafa Soleiman says the goal is to give co-pilot a lasting identity presence and even a digital room where it lives and ages. So this marks Microsoft's latest attempt to humanize its digital assistance following earlier attempts from decades ago, like the animated clipy, which was often criticized for being intrusive. So right now, the feature is only live for select users in the United States, UK and Canada, with no timeline yet for a global rollout aside from being available online via Microsoft co-pilot inside copilot laps. What do you guys think?

You guys excited to see a kind of an AI version of Clippy make a comeback. I don't know. For me, at least when it comes to interacting, you know, with different AI, you know, voices. assistants, I don't need the cutesy animations, the drawn out. That's me. I know some people like it. You know, if I'm talking to a voice assistant, I want it to be direct. I want it to be fast. I don't want it to be, uh, you know, overly verbios and, uh, you know, give me a 900 word answer when I'm looking for nine words. So at least for me, uh, the, the new co-pilot appearance, probably not something I'm going to be personally benefiting from. I could.

could be wrong right as it starts to roll out, you know, kind of this aging thing in where it better understands your conversational history. But at least on the surface, I'm not here for it. Maybe people are. I don't know. Maybe people miss Clippy. All right. Our next piece of AI news, which I've been saying has been coming for a very long time.

Netflix has announced that it used generative AI for the first time in one of its blockbust original scripted TV shows. So the milestone happened in its new original series, L. Eternuta. Hopefully I got that right. I haven't watched it yet. So it's L. Eternata.

I think that's how it's pronounced. We'll see. In Argentine science fiction series on Netflix, where AI powered tools helped them create a building collapse sequence in Buenos Aires. So according to Netflix co-executive, CEO Ted Serendos, the AI-enabled visual effects sequence was completed 10 times faster than it would be with traditional visual effects methods. So the show itself featured a staggering 2,000 virtual effects shots and its visual effects

quality generated significant buzz after the series premiered in late April. So Netflix claimed that AI will only be. be used to lower cost and speed up production, but also make advanced visual effects, like de-aging accessible to more creators, not just big budget projects. So Netflix has previously faced criticisms for its AI use, including AI manipulated images in documentaries and reconstructing the voice, reconstructing the voice of a murder victim. So the use of AI in film and TV remains controversial with ongoing concerns about job security for actors and artists as well as the artistic integrity of AI generated content.

So other streaming platforms such as Amazon are also experimenting with AI, using it for show recaps, content recommendations, and voice debugging. So Netflix has also, according to Bloomberg, started using video generating AI from the startup runway for content production, although that tool was not used in this case in El Eternata. So, no, I've been saying this for more than a year. AI generated video is going to be the norm, right? And we're going to start seeing it without even knowing, which is in this case. So actually, the visual effects for this show have gotten positive reviews, right?

Which is why a lot of people were kind of up in arms when it was revealed that Netflix used generative AI video tools to help create some of these visual effects scenes. y'all, I think it started, honestly, with the rollout of certain video features in Adobe's platforms where you could use generative AI to extend a scene by four seconds. But I do assume that by this time next year, a good percentage of the TV shows and movies that we watch on the big screen are going to have elements of, generative AI, right? I don't think, you know, we're going to have entire blockbuster films created by AI, but it's going to be commonplace for certain segments, maybe even entire scenes to be AI generated. And most users are not going to know the difference because AI video has gotten so much better recently. And if we're being honest, if you compare it to text

generation, it's a little expensive. Right. So this is one of those areas that the big players are actually able to monetize, where we've seen reports when it comes to AI chatbots, some of these big AI labs are actually running them at a deficit. But the most powerful AI video tools in the world are a little bit more expensive. So I do think that they've kind of found a good product market fit here. And I do see that continuing to be the trend over the next couple of years. And I've said this, you're going to be watching entire AI shorts. I don't think, you know, by this time next year, we're going to be watching entire, you know, blockbuster films that are AI generated, but they will all have AI generated scenes, AI generated effects, essentially when it costs, you know, a company millions of dollars or multiple months when they have to go reshoot something, right, for a couple, you know, 10 different 10 second scenes. They're just going to use AI, right? And the majority of audiences are not going to know the difference.

All right. Our last piece of AI news. The AI driven fraud crisis is coming. So Open AI CEO Sam Altman has just issued a strong warning about the risk of AI driven scams and impersonation speaking at a federal reserve event attended by major U.S. financial institutions. So Altman told the Federal Reserve that AI has already defeated the most current authentication methods except passwords and called it crazy that some financial institutions still rely on voice

authentication to move money because AI can literally copy anyone's voice without any needed authentication, right? Anyone in seconds can copy and clone anyone's voice on the internet, right? It's not exactly legal to do, but it's extremely easy. So Altman warned the group that the world is. on the edge of a quote unquote significant impending fraud crisis with AI powered voice and video impersonation scams already targeting parents and even high level government officials, which we shared about on the show a couple of weeks ago.

So the FBI has previously cautioned about these AI cloning scams and U.S. officials recently reported in AI generated voice impersonating Secretary of State Marco Rubio in attempts to contact foreign ministers and politicians. Yeah, it's going to happen. And plenty of people are going to be duped by this, I would say, in the next year, until it is really widely known how good AI voice cloning and even video cloning is. Right. I think we live, if you're listening to this podcast, watching the live stream,

I think we kind of live in an AI bubble and we just assume that everyone knows that anyone can go clone anyone's voice in seconds and make it sound indistinguish. from the real person and the same thing with video. But here's the reality. 99% of the global population has no clue that this technology exists at the level that it exists now. It is very hard to tell the difference between what is real and what is AI generated. So Open AI is expanding its policy presence, announcing a new Washington, D.C. office to work with the federal government on AI policy, train government officials and research the economic

impact of AI. So despite these risks, Open AI has urged the Trump administration to avoid regulations that would make it harder for U.S. companies to compete globally, especially against China. So yeah, you kind of have these two competing storylines here, right? So Sam Altman essentially warning federal officials that these AI generated fraud is going to lead to a crisis. Yet at the same time, The White House essentially gave big tech a free pass to build AI however quickly they can. So part of it makes sense on the surface, right? Regulation is going to slow down the global AI race. And let me tell you this.

The global AI race, it's not about the military. It's not about, you know, precious materials, gold, silver. It's not about oil. It is about AI. It is about the most powerful chips, number one, and the most powerful models, right? That is the future of global superpowers is whoever wins in AI, whoever achieves artificial general intelligence, artificial superintelligence first, gets an unfair advantage for probably years or decades when it comes to being a global superpower.

So pretty interesting story here because Sam Altman on one side, is trying to warn, rightfully so, the federal government that we are going to see an impending AI-driven fraud crisis. Yet at the same time, all the big tech companies don't want to be regulated. And it looks like the White House isn't going to regulate them. So what this is ultimately going to lead to is a pretty huge AI fraud crisis. So Altman's comments came, like I said, at the same time as the White House unveiled its AI action plan and Altman has voice concerns about bad actors misusing AI super intelligence for

attacks like targeting power grids and creating bio weapons and about society losing control over powerful AI systems. So on the topic of jobs, Altman said that while entire classes of jobs will disappear, new work will emerge, though he believes no one can accurately predict the full impact of AI on the workforce. All right. Let's do a quick recap now of our new kind of rumors and rants or kind of what's new and what's next. Yeah, there was a ton announced that didn't make our kind of big AI stories.

So here they are. All right. So Open AI has announced a new deal with Canvas, the AI educational tool, which is a little confusing considering it has its own canvas mode. Microsoft co-pilot vision will be. rolling out in select Windows 11 PCs, so not just on the web. Google Web Guide, a new way that AI search is grouped together has just been announced, and Google is starting to test it.

Runway has announced ALIF, hopefully that's how it's pronounced, a new video editing tool for multiple task visual generation. So developing or sorry, creating multiple AI video scenes at once. Google has started to roll out its new AISC, powered try-on shopping feature. So much Google stuff. Google Photos has released a new AI remix feature. ChadGPT is starting to roll out a new personality feature,

at least on the iOS app, that kind of predetermined ways that ChadGPT will respond. Microsoft co-pilot's new research is now editable. So inside Microsoft co-pilot, they have their version of deep research before the plan was, not editable. Now it is. And it also now is going to show the chain of thought summary, which is pretty cool. XAI's Grock is getting a new task automation in app connectors soon. One I'm very excited about. Notebook L.M has started previewing. So we've seen the first actual

videos that Notebook LM has produced, which I think are going to be insanely popular, just as popular as the Notebook LM podcast are. Claude may soon begin. getting temporary chats, file downloads, and memory. And now Claude Code allows for creating subagents that can work together. All right. That was a ton. Let's quickly recap the major AI news stories that matter for the week of July 28. So OpenAI is set to reportedly launch GPD 5 in early August, according to the verge.

Google has launched Opel, a no code AI vibe coding tool. InVIDIA B-200 AI chips are flooding China despite a U.S. ban. Reportedly, at least a billion dollars worth of ban chips have made their way into China, according to reports. U.S. President Donald Trump has signed a sweeping AI action plan, prioritizing U.S. Tech and D regulation. Microsoft's GitHub has launched their AI-bidecoding tool, Spark. Then we had Google pursuing a news licensing deal with more than 20 national news outlets.

Microsoft is bringing back clipy, kind of, with their new co-pilot's appearance, which gives co-pilot a face and visual avatar. Netflix announced that it used generative AI video tools for one of their new hit series, and OpenAI CEO has warned the federal. government and banking institutions of a looming AI-driven fraud crisis. That was a ton to cover. So like I said, we do this every single Monday with our AI News That Matters segment where we cut through the fluff, just tell you what's new, what you should be paying attention

to, you know, don't give you a bunch of BS. So I hope this show was helpful. And then we do this every single day, actually, not the AI news that matters. But on Tuesdays, we do our hot take Tuesday. on Wednesday, we put AI to work on Wednesday going over practical use cases of new generative AI tools. And then mostly on Thursday and Fridays, we interview some of the brightest minds and leaders in AI.

So that's kind of our weekly lineup that we've settled on over the last couple of quarters. So I hope this is helpful. If it was, please share this on social media. Don't just keep everyday AI as your secret cheat code. The only way we can keep this thing. going is when you tell people about it. If you're listening on the podcast, we appreciate your support. Please make sure to like and subscribe to the channel. Leave us a rating. We'd appreciate that. And make sure after you're done doing that to go to your everyday AI.com, sign up for the

free daily newsletter. So, thanks for tuning in. We hope to see back tomorrow and every day for more everyday AI. Thanks, y'all. Meet Firefly AI assistant. Now live in Adobe Firefly, the Allman One Creative AI Studio. Just describe what you want to create in your own words and the assistant handles the rest, orchestrating multi-step workflows across Adobe Creative Cloud apps, including Photoshop, Premiere Express, and more in one conversational interface. You direct the outcome while the assistant accelerates execution. Stand control with the ability to step in and refine at any time. See it today at firefly.adobie.com.

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