Loading summary
Transcript1 lines
- [00:00]
A
We have some major news out of Google Gemini. Their AI model has gotten a whole bunch of really powerful upgrades. They've added something called Gems, Assistants, and of course, Imagen 3. So a ton of big updates to talk about. Today. I'll be covering all of the podcasts. Before we get into it, I wanted to let you know if you have ever thought of starting a podcast or you have a podcast and you grow it bigger, you need to take my podcasting course. I've recently launched a pod course that takes you through every of starting a podcast from start to finish. Everything that I have learned in the last five years of podcasting, I have made a ton of mistakes and most of them I would not wish for you to have to repeat. There's things that I've learned I've spent, you know, tens of thousands of dollars on gear, equipment, building an entire podcast studio and production team for my podcast and for a lot of the stuff that I work on. So if you have ever wanted to start a podcast, grow it. I have a course that teaches you everything I learned in starting my podcast, how to grow from zero, and how I got to 4 million downloads, was able to raise hundreds of thousands of dollars, grab thousands of customers for my businesses. Everything that I've been able to do with podcasting, I just think it's one of the best mediums for growth, marketing and setting yourself and your network apart. So if you're interested, there's a link in the description. Normally it is $300, but for this week I have set the price to be 50% off with the discount code AI Chat. So there's a link in the description where you can use that. So if you've ever thought of starting a course, a podcast, I would grab that link and join the course. I think that this will be something that you that is amazing and really helps you in that regard. So let's get into what Google Gemini is doing. First thing I wanted to bring up here with everything that they've announced. They had a whole Tuesday announcements on some really new significant updates that they've been doing. Logan Kilpatrick over on X, he's, you know, works with their head developer relation team and he's been tweeting out a bunch of really amazing updates. The first thing I wanted to talk about is something that they've introduced called gems. So gems over on Google Gemini essentially is one of their latest innovations that I think is quite interesting. Essentially it's custom AI experts or kind of like agents, I guess. So you can have These pre made gems essentially it's kind of like chat GPTs. Their GPTs you have desk researcher, code reviewer, French tutorial. So they call them gems, they're essentially like experts. But it's a lot to me like it's a lot like the GPTs that OpenAI has. And you can essentially start with these pre built gems or these pre built cases and use them to kind of start working on things right away. You can search for them and find them or you can just, you know, go and scroll around and kind of get some inspiration on things you would want to use. ChatGPT or I mean, I guess, sorry, Gemini for now also very obviously famous for their enterprise use cases. This is something that they've put a big focus on especially with Logan Kilpatrick over there working on developer relations in some of the startups I'm looking at. They've, many of my developers have told me that Google Gemini is one of the easiest platforms to actually work with. Integrate with their API is very straightforward and they have went through great lengths to add, I don't know, very, very developer friendly features to this. So like massive context windows, millions of tokens, free tokens, all sorts of, yeah, all sorts of really powerful technology that usually be very expensive for a company to run. But it's Google so they're kind of throwing everything at it, trying to make this really, really popular. So right now I think this big kind of shift towards the specific kind of AI assistance or these gems that they have is interesting. It's, it seems like it's kind of in response to what OpenAI is doing. So I think that's very exciting. But the other thing I think that's very exciting is their Imagen 3. So they're upgrading their image generation capabilities with this latest version. It's currently available to all users of Gemini and it has higher quality image. You're essentially just, you know, using text, prompting it, getting an image. And they are also including human image generation, although they do have some restrictions. Previously I think they're, you know, they weren't quite. Well, they had a lot of controversy with previous image models and generating historically inaccurate images. So if you asked it for, you know, a World War II soldier, it was going to generate, you know, a World War II German Nazi soldier, it generate like a black person and obviously that never existed back then. So it got a lot of controversy flack and you know, went viral for a lot of reasons and I think because of that they kind of pulled some of the image generation capabilities for humans now that seems like they're, they're bringing it back, but they do have some restrictions, so. Yeah. Anyways, I think this is very interesting. In order to essentially address concerns about deep fakes, Google has implemented safeguards. They have something called Synth ID watermarking technology that is going to essentially kind of watermark so you can know if an image is real or not. The effectiveness of this though, we don't really know. There's a lot of debates going on around this, but it'll be interesting to see how effective that all plays out. It definitely is a crowded market right now. Like Google is not releasing this without any competition. There's been, you know, very similar products and things coming from OpenAI, Microsoft, Meta anthropic hugging faces. They are all building kind of these customizable AI chatbot platforms. So I think that, you know, their new gems or their GPTs are not completely a new thing. Imagen3 I think is really where they're hoping to catch up and maybe pass some of their competitors. Not everyone is doing image generation, but that being said, you know, even companies like xai, you know, on Twitter using on like Grok, their AI image thing over on X, they don't have their own image generator, but they were able to grab open source image generation platform and embed it into Grok and essentially use that to, you know, get image generation right in their model without having to develop it themselves. So Google's trying to, you know, get past everyone and come up with something new and unique by having their own image generator, which is fantastic. But I just think that some of these open source ones are so good right now that I don't really know if it's actually giving them that much of a competitive advantage because all of their competitors can essentially just do what XAI did and grab some of these open source models that are powerful and then they immediately have image generation inside of their text models as well. So I think that's going to be really interesting. Google is definitely making a big push. I'll keep you updated on everything that they roll out. Um, as far as the quality goes overall, I think that, you know, I would not count out Gemini, especially with some of the developers they have over there. Logan Kilpatrick is doing a phenomenal job. If you're interested in what's happening over at Google, I'd follow him on X to see what's going on. And again, if you are interested in starting a podcast, check out the link in the description. I'd love to have you as a member of the course. And this week is a 50% off discount. So if you're interested in doing it, now would be the week. I hope that you all have an amazing rest of your day.