Deep Dive
Shownotes Transcript
Welcome to the AI Chat Podcast. I'm your host, Jaden Schafer. Today on the podcast, we're talking all things Google. There's a ton of big updates with Google AI, and they have essentially rolled out a new investment fund where they're going to be investing into AI startups.
They've done a couple things in the past, but this is one of the biggest moves I think they've made to date in this direction. So we'll be covering that. And we'll also be breaking down the latest numbers we just got today from Google Gemini usage. So this is really fascinating. Recently, I did a podcast talking about the enterprise adoption of a lot of AI companies and Google Gemini.
really got hammered. They, uh, went from like 2.6% down to like 0.1% of enterprise adoption. But just because the enterprise is struggling does not mean the consumer side is struggling. They have some really interesting new metrics, uh, that we're going to be getting into. And before we do that, I wanted to say, if you've ever wanted, um, to get access to all of the top AI models on one platform and ditch, you know, your subscriptions to Anthropic, OpenAI, um, um,
Gemini, all of the other AI models, you can check out my startup AI Box's playground. So it's AIbox.ai. And on our playground, you have access to all of the top AI models. You can chat with them all in the same thread. You can rerun, you know, if you have a question, you can rerun the same question on seven or eight different AI models to see which one responds best. You can compare responses side by side of different models you have. This is really cool, especially when you're doing audio or image, I find.
Um, and you also have something called media storage. So every piece of media, audio, or image you've ever created will show up in your storage. You can then go and click on your media, see what exact prompt you use, see what model you use, how many tokens. And the thing that I love about it is that you're able to go and click right back into the chat that you're using. Um,
when you generated that image or piece of media. So like sometimes I'll be creating a thumbnail for a podcast or episode and I'm like, where the heck did I like create that? Because I want to tweak it and use it again. Pretty much I go to the media storage, find it, click it, go back to the original place and then upload whatever I want to make some tweaks. So really cool. You can check it out. AI box.ai. It is $20 a month and saves you on subscriptions to a ton of other platforms. All right. Let's talk about what's going on at Google.
So they have a new initiative. Essentially, they're going to be backing AI startups. And they just announced this on Monday. Now, the interesting thing here, specifically who they're going to be backing, is any AI tools that essentially startups that are building AI tools that are using Google DeepMind, which is pretty much the company's AI R&D lab. So pretty much, you know, they want people that are using Gemini. They want people that are using...
any that are using their own tools to build things. And it's called the AI Futures Fund. So it's going to essentially back anyone from seed to late stage. They'll have a whole bunch of different support. They're going to allow founders to have early access to some of the Google AI models from DeepMind, which is kind of a cool perk. Assuming Google can keep up, I think, like if I'm being honest, if Google can keep up with all the other top
models, that's fine. That's cool. Getting early access is always awesome. But when Google's early access might be behind open AI or, you know, anthropic, whatever they have on the market today, it's less cool. So I think this is, this will be awesome if they can, you know, keep up with everyone. This is going to be great for startups. So right now we have, you
kind of that is one of the big features. We also have the ability to work with Google experts at DeepMind and Google Labs. That is a huge asset. And then of course you get Google Cloud credits. So when you're running, you know, compute on these, or when you're running these AI models for whatever your startup does, getting those credits is essentially like getting cash because you've got to spend a lot of money on this stuff. So
Some startups are also going to have the opportunity to receive direct investment from Google itself, which is pretty cool. This is what they said. A spokesperson over at Google said, they said, quote, the AI futures fund doesn't follow a batch or cohort model. Instead, we consider opportunities on a rolling basis. There's no fixed application window or deadline. When we come across companies that align with the fund's thesis, we may choose to invest. We're not announcing a specific fund.
fund size at this time. And check sizes vary based on the company's stage and needs, typically early to mid-stage with flexibility for later stage opportunities as well. That's kind of funny because it's kind of like saying we invested pretty much everything. But I think really their focus is going to be on early stuff, but they don't want to shoot themselves in the foot if some late stage thing comes down the pipe that's really promising. They're like, well, we'll also check that out too. But I think the goal and the big focus will be early stage stuff. So
AI Future Fund already has a couple case studies. There is an AI meme-making platform called Vigil, and there's also a webtoon app called Toonstrata that have participated in this program. And startups can actually start applying as of today to join this. So
Google has been making a whole bunch of big, I think, promises in regards to funding AI stuff in the last little bit. In November, the charitable wing of Google said that, so that's google.org, they said that they were going to have $2 million as a cash commitment to researchers and scientists, anyone that's kind of working on AI from research and science perspective.
And then in September last year, the CEO of Google, Sundar Pichai, said that the company was going to create a $120 million global AI opportunity. This was more for education and AI and education and training and kind of all around the world. Google.org also launched a $20 million generative AI accelerator program. And that was pretty much giving checks to nonprofits that were developing AI tech. So Google's been doing a ton. They've been giving out a lot of money. But this, in my opinion, is kind of like
the first big step into just generally investing into AI startups. I think they know that this is an important thing for them because if they're not able to do this, a lot of other companies are going to be invested in them. Microsoft, OpenAI, Anthropic, a lot of these big AI companies, a lot of their quote unquote competitors are putting money into other AI startups. I think this is a good thing as Google feels like they might be
getting a little bit behind as far as consumer adoption goes. They want to invest in startups that are going to help push their usage of their tools. I think this is a smart move.
The Google for Startups Founders Fund is another thing that Google has I just wanted to mention, but essentially that just supports founders with a bunch of different backgrounds, including AI, but it's not specific to AI. It's just kind of any startup. Although, to be fair, all startups today seem to be AI or have some sort of element of AI in them. So I think that this is kind of like...
You could say that this was an AI-focused fund, but in reality, this is just a general startup one. So now we have kind of the official, the first official AI startup fund. And I think we'll actually start seeing Google make some big plays here. They're going to want to make some announcements. That's super, super exciting. The other piece of news I wanted to mention around Google is in regards to their own technology, which is really, really fascinating. And that is that Google's Gemma AI model has just passed 150 million downloads.
So this is essentially the open source model that OpenAI has released. It's kind of a lightweight model. It can run on things. And it essentially kind of competes with other quote unquote open AI models like Meta's Lama.
And this is really impressive. 150 million downloads is absolutely incredible. Now, the one thing that I will say is while that's like really impressive and exciting, it doesn't actually directly benefit or help Google in regards to their Gemini usage. So it's exciting that people use their open source model.
But the Gemini usage hasn't really seen a direct benefit to that. And so Google's slightly struggling there. So hopefully, I think the kind of the idea here, though, is that people are going to be building with Gemma. And since it's a lightweight model, lighter weight model of Gemini, hopefully it kind of moves people if they're like, hey, I need it to be better or faster. They can kind of plug it into a plug in their current stack into Gemini. And maybe that will help them. Yeah.
Llama though, if you're wondering like how it compares this Gemma model to Llama,
Llama has 1.2 billion downloads as of April. And so 150 million is really impressive, but it's only a tenth of what Meta was able to do so far with Llama. So I do think this is interesting. And I also think it is worth noting that Gemma as well as Llama, they both have been criticized for these kind of custom non-standardized licensing terms. Some developers are saying that
these have essentially made using them for commercial use like really risky. It's kind of like, it's kind of like they're like, hey, you can use them. But if your company is this big, you got to come to us. And so a lot of people just aren't using it unless you're kind of like these indie devs or doing smaller projects where you don't think the
come after you. A lot of interesting things. In any case, if you enjoyed the episode today, make sure that you go over and check out AIbox.ai. You can test out the latest models from OpenAI, the latest models from Google. We have a ton of really interesting stuff with Google Gemini Pro 2.5 and others that you can go test out on our platform, AIbox.ai. Go and check it out. Thanks so much for tuning in to the show today. I hope you have a fantastic rest of your day, and I will catch you next time.