00:00Hello, everybody, and welcome back to another video where we talk about everything in tech,
00:03data, and AI. And in this week, we have a lot of things to cover. And interestingly enough,
00:09we have a lot of things about Google. The first one being that the CEO of Google just announced
00:13that the Gemini app, which is their AI app, just reached 1 billion users. Now, this is their 14th
00:20product to hit 1 billion users. Now, I am not saying that that is not a huge thing. But what
00:27they know that have 1 billion active users. If we look at this article talking about this
00:33milestone, it says right here, the company has steadily been integrating Gemini across its
00:38products from search and workplace to Android and its standalone app. Google's AI mode in search in
00:44particular has achieved a lot of success with over 1 billion monthly active users globally. Now, again,
00:51I personally use Google. I like Gemini. I think it's a great product. But I'm very curious to see
00:57how they're actually doing this. Does this mean that anybody who goes onto Google, like google.com,
01:02and asks a question and gets their AI kind of description that they are just known for now?
01:06Does that mean that they are an active user? Because that just means everyone who's Googling
01:10anything is going to be an active user within these parameters. So I'm not exactly sure how they
01:16were measuring this. But one other thing to note is that it also says that Google also shared numbers
01:21of how people actually use the chatbot with 63% of Gemini users talking directly to the assistant
01:28using the voice feature and that they're generating over 150 million images every single day. 63%
01:36is incredibly high. I mean, I am honestly pretty shocked by that. I would have guessed it would have been
01:4220%. But 63% is very high. So if nothing else, it's super interesting to see how people are actually
01:50using AI, how people are using Gemini. But purely from a data standpoint, I'm just wondering, are people
01:56actually going out of their way to use the AI functionality within Google? Or they just happen
02:02to be using it or using the system that has AI integrated into it, and they're counting that as an
02:08active user? I mean, as a genuine question. The next thing that Google is in the news for is right
02:14here. Google wins bankruptcy auction for Spirit Airlines emails, chats, and documents. That's right.
02:19The airline company that went underwater recently, they are selling off bits and pieces of it. And
02:25Google won this auction, and they won it for what I would consider a very small amount. It says right
02:30here that Google won a bankruptcy auction for the data, agreeing to pay $10 million for the information.
02:36Now it says the deal includes 100 million emails and 500 million Microsoft Teams chats, according to a
02:44court filing. Now what are they going to use this data for? You guessed it, they're going to be using
02:49it for AI training. It will use the data to improve its products and AI models, which right now, I
02:56mean,
02:56they're just really pushing their AI stuff. They're trying to train their models. They have their Gemini
03:003.7 out right now or whatever it is. And so they're going to be using that data to train
03:05their models
03:05for sure. Interestingly enough, they did specify in here, because I was curious about this myself,
03:10the data does not contain any personally identifiable information. They're going to
03:15basically de-identify this data before they use it. Yeah, it says it right here, meaning it won't be
03:20associated with individual people. And the buyer agrees not to attempt to re-identify the user. So
03:25Google can't go out here and say, okay, this person, Alex Freeberg, is a Spirit Airlines customer.
03:31They can't know that information. They shouldn't know that information. I think $10 million is an
03:36absolute steal for this data. I guarantee you that Google is spending billions of dollars elsewhere
03:41buying data from a ton of other places. This is probably a small percentage of all the data that
03:47they're collecting. But I mean, this is really useful information, especially that 500 million
03:52Microsoft Teams chats, because they want their AI to sound really natural and really human. And that's
03:57just a great thing to train their data on. So I'm not surprised that they were interested in this data.
04:01It's just interesting that Spirit Airlines went under. And now here's Google, you know, buying the
04:06leftover pieces and scraps of data. Believe it or not, we have one more story on Google this week. And
04:10this mostly affects developers or app developers. It says Google buried rival app stores in their
04:16Play Store, and Judge gives them a week to fix it. Now, this is very recent. You can even see
04:21right down
04:21here. They have to fix this by August 20. And there are three specific interface changes or UX changes
04:28that they have to make by that date. It says that the court found specific Play Store behaviors
04:32anti-competitive and Google's own counsel agreed to all three corrections on the record. So at least
04:38they're like, hey, yeah, we know this is an issue. We probably shouldn't be doing this. And we'll make
04:43those changes. If nothing else, that's a good thing. But the fact that they're, you know, putting these
04:48out there allowing these changes to be made to where specific apps just aren't gaining any traction
04:53even by big players. And you can look into this. I'll have this, you know, down in the description
04:57if you want to check this out. It's just really interesting to me. And so some positive things by
05:02Google, some, you know, not so things by Google. But Google's a very, very, very large entity. And so
05:07again, there's going to be some good with the bad. The last story we have is about Meta. And this
05:13article
05:13talks about it right here. It says, states take Meta to trial in California in the biggest fight yet
05:18over social media harm to children. Now, this has been a big thing for a long time, right? There's
05:23a lot of negative effects to social media, especially for kids. And what they're doing
05:28on California right now, I'm actually a big supporter of. I mean, there's a lot of harm that
05:32can come to children. And they are, at least in theory, totaling as much as 1.4 trillion in damages.
05:38And that's a huge amount. And to no one's surprise, it is about mental health crisis and addiction.
05:44It says the lawsuit accuses the social media giant of contributing to the youth mental health
05:48crisis by knowingly and deliberately designing features that get children addicted to their
05:53platform. It also claims that Meta routinely collects data on children under 13 without
05:57their parents' consent in violation of federal law. Now, we could really dive into this, but
06:01I think you guys are really well aware of how a lot of these things have played out in the
06:05past, right? They get a slap on the wrist, they get fined, and then they keep doing it because
06:11they make so much money from this and they kind of get people addicted in social media to where
06:16they do it for the rest of their lives. And so they're willing to pay out however much money
06:20they end up paying out. I doubt it's going to be this 1.4 trillion. My guess is it'll be
06:25like,
06:25you know, 50 billion, which is a lot of money. But to these big tech companies, that's nothing.
06:31That's just like their marketing budget for a quarter four. And so it's no big deal to them.
06:34They're going to be, you know, A-OK, you know, whatever they get fined.
06:38For our very last story, have you guys heard of Metabase? I have. It's a really cool platform.
06:44It's an open source AI analytics platform where you can basically build out semantic layers,
06:49business logic, and all these different things. It's all open source. It is really cool. And it's
06:54super huge on GitHub. And of course, they have like a paid option that you can get a lot more
06:58features for. But this is all open source, which is really, really cool. The unfortunate part of this
07:03is that Metabase is in some really hot water over some zero-day exploits with none other than SQL
07:09injections. It says right here, a critical Metabase SQL injection vulnerability was exploited in a
07:15zero-day attack to breach customer instances in data theft attacks known to Impact Framework and
07:21Tally, which are two separate companies. The funny thing is that SQL injections are not new. And
07:26honestly, you should be able to really easily not allow that. But somehow they did. And so people
07:33are like, hey, SQL injection, that's like the first thing you try when you have access to API
07:38endpoints or different things where they grant you access to run code. It says right here,
07:43the company says attacks can be identified by a post request to API session reset password,
07:48returning a 400 status, followed by a successful get request to API user current. And Metabase warns
07:55that the system log showing these entries have likely been compromised. In my humble opinion,
08:00and again, I'm not in cybersecurity, but I did work in IT for analytics. This should not be happening.
08:05This shouldn't happen at any company. But it does happen. It still happens. And SQL injections are
08:11something that you should be aware of, especially when you're giving access for someone to write code
08:15or to write anything into a database or any window where you can write code. They have now since patched
08:21this. But of course, there are affected users where all their data got taken. Honestly, this is nothing
08:25new. It's just really interesting that companies are still getting hit with SQL injections,
08:30and they're still working. That's the shocking part. So with all that being said, that is all we have
08:34for this week. Again, we have a lot of Google stuff. They were just in the news a lot that
08:38I was
08:39personally seeing. And I am not shocked that they reached a billion users. In fact, in this gif that
08:45they used right here from Sundar Pichai, I am not shocked that that is, you know, an AI image.
08:52It looks like an AI image. But one billion users is a huge amount. That's a lot of people. I
08:59mean,
09:00that's like, you know, an eighth of the world. It's a lot of people. And so I wonder how many
09:05are bots? How many are real people? I wonder how they're collecting that data, how they actually
09:10define active user, right? There's a lot of things that have questions on. And, you know,
09:15I guess we just won't know. They just have a billion users. But it is really impressive.
09:20Nonetheless, that is a lot of people and a lot of people using their apps and their platform.
09:25With that being said, thank you guys for joining me today. If you like this video,
09:28be sure to like and subscribe. And I will see you next week.
