Aria Unleashed: Opera's AI Assistant Makes Its iOS Debut

0:00In the world of artificial intelligence, one of the most controversial issues is that these massive language models are monetizing essentially your data by sucking up all the data that they find out of artificial intelligence. is planning on doing, what their strategy is, and why I think this is more important than ever in the world of AI. So let's jump into it. I think in a world that is increasingly conscious of online privacy, a rising tide of startups is really kind of focusing on enabling individuals to monetize their data. The cadence is obviously one, but there's other companies too, right?

0:49There's Datacy, which allows users to collect and sell their browsing data to the highest bidder, while platforms like Pogo and Tapestry offer rewards in return for access to behavioral and purchasing patterns. But this newcomer that I've mentioned today is called Cadence and this one's kind of setting itself apart with a bit of a unique approach. So Cadence essentially grants users intricate controls allowing them to decide exactly which data they wish to share with third parties. So Cadence model essentially involves aggregating analytical data on users movie preferences, travel habits, shopping behaviors, and more.

1:26And users then have the option to monetize this data in various ways. So Caden has the backing of a really noteworthy name, that's Jeremy Yang, who is the founder of Yahoo. And the concept for Caden was sparked by John Roe's observation on the evolving online privacy landscape. So Roe, who successfully sold his private startup act to Salesforce in 2015, noted in an email interview with TechCrunch that he recently did, he said, quote, for 25 years, users have exchanged their personal data for free services and access to apps and websites. I envision a future where users would be at the heart of this data exchange rather than being passive participants.

2:08This is really interesting. And actually, I think it goes like one other anecdotal or interesting antidote on this is the fact that Facebook tried doing a play a while back, if you remember, where essentially they had an app that was on iOS. This was shortly after iOS kind of banned Facebook from tracking what users were doing across their phone. If they weren't actively using Facebook, Facebook wasn't able to essentially track everything they were doing on their phone as they previously were able to. And in response to this, Facebook, I'm not sure if they started or if they acquired essentially a company that would give you, you know,$20 gift cards and rewards to allow them to track you.

2:48And there was a bunch of different like stuff you could do there. iOS did not like it kicked the app off of the app store. There was a whole debacle about it, right? This was kind of in the Facebook versus Apple feud that went on around privacy. Now, it's interesting for me to see kind of a whole new version of this, where essentially, these startups are allowing you to monetize your own data, but Facebook kind of was letting you do that before, right? They would give you like little rewards. There's a couple others doing it. So I'm curious to see what the big differences here are and if these will succeed where Facebook previously failed in this regard.

3:22So I think one thing to take note of is the fact that privacy laws definitely have multiplied. In 2023 alone, almost 140 consumer privacy bills were introduced or contemplated across over 25 states, including Puerto Rico. And as per the National Conference of State Legislatures, that's where we got all that data from. But I think the other thing that's really interesting is that while international legislations like the EU's GDPR, and then there's also some domestic laws like the California Consumer Privacy Act, which essentially are set up to protect users' rights against unwanted data tracking and sales, they don't actually directly offer avenues for users to monetize their data.

4:04So ROA recently emphasized Caden's uniqueness in this space, saying, quote, Our core offering empowers users to control their data, rewarding them if they share it with a trusted third party. So the Caden experience essentially just starts with users linking their online services, right? So they can link their Amazon, their Uber, their Netflix. And they do all of this through Caden's iOS app. and users can view statistics, insights, and even opportunities to monetize specific data segments by sharing them with advertisers. So of course, I know the question on everyone's mind is like, you know, data, is there data privacy?

4:44Are they like literally knowing exactly everything you do? Is there a profile being made on you? All those kinds of issues. I think given the sensitivity of the data that is being shared, security is really important. So ROA says, essentially, they've told users that all of the data is secured within an encrypted quote-unquote vault, which, I mean, whatever. They just throw the word vault in there to make you feel more secure about it. But, I mean, allegedly, it's encrypted, so that should be good. The thing that I find even more important, though, is that it is anonymized before third-party sharing, right?

5:15So, they're not going to know this is exactly you or one specific person sharing this data. They just know that, in general, people with your characteristics, a person with your characteristics has these habits, these shopping habits, these travel habits, etc. So plus users also have the ability to retract sharing permissions or even delete their data at will, which I think is important in this space. Now, would I do this? I don't know. I guess it kind of depends on how pinched I was for money and how much I could get paid for it. Probably wouldn't be my personal choice. I'm just a bit more of a privacy nut than that.

5:53But I do see how this has appeal for people, and I do see how this, you know, it could be an interesting play. So reportedly, Caden has already amassed over a billion data points and has compensated users by tens of thousands. So they're further incentivizing participants with the introduction of Caden AI, which is an artificial intelligence assistant that makes tailored content and product suggestions based on shared data. Roa believes this offers a more personalized experience for users, allowing them to discover products and deals tailored to their tastes and budgets. Now, that is obviously an interesting feature.

6:30But the reason I really bring Caden up on this podcast is because of the implication I see in artificial intelligence. So right now, really what they're doing is they're saying, hey, share your data with us, we'll send it to a travel sites and advertisers and some other people, blah, blah. But where I really see this going is this is going to be a massive boon, whether it's Caden or another company that is like them, for AI startups. Inevitably, what's going to happen is platforms that monetize your data are, I believe, the gatekeepers right now, but I believe the power will eventually get put back in the hands of the people where you will be able to sell your data directly to, through, you know, an app or a third-party marketplace like this, to companies like OpenAI and other people that are training AI models off of your data.

7:18So beyond just right now, the big play that everyone's talking about is advertising. I think this is kind of like if you're looking at the gold rush, the data for advertising is a much, in my opinion, is probably less lucrative than some of the AI, the data used for AI models. And so I think that consumers could actually perhaps sell it twice, advertisers and AI models, but I think they can make a lot of money. And I think AI models would be very wise to participate in these kinds of schemes and programs because essentially anyone, even if you don't have access to a massive database, right? You can train a very powerful AI model.

7:56And I think that's one of the big problems with, you know, a lot of startups today, you know, everyone makes fun of all of these startups that essentially are just a wrapper on top of an API access to open AI, right they're just like you know write emails write articles write xyz thing with access through open ai's api and it's like you know they don't have a very big moat they're easy to replicate etc etc the the value in the brand is just in the marketing and the users they can get to use their product but if they were able to actually train ai models to do some of the things that they were doing and create custom models that were better than open ai and could do things you know more if they could do things that were a lot more powerful there would be a lot more value in their companies and they would be able to charge more to consumers and potentially grow more and so I think the big problem though is that they do not always have the data to do that and so there's some places you can buy data from but this is definitely one way where you can get access to data without having the controversy of consumers saying hey like I didn't give you permission to you know we didn't give open AI permission to scrape the entire internet so I think it's going to be interesting to see if these kind of data, you know, pay for data plays directly to consumers takes off in a sense that AI models are going to be going to be buying them from these kind of platforms.

9:10On the business end, Caden is developing Context AI, which is aimed at aiding enterprises, especially marketers and understanding consumer behavior. And this tool is going to answer natural language questions and provide visual data representations. Investors are really bullish on Caden's trajectory. I think a recent announcement revealed that Caden raised their Series A, right? It was$15 million. And this round included people from Neva Ventures, AME Cloud Ventures, Streamlined Ventures, Montage Ventures, and a bunch of others. So I think changes in industry practices like Apple's app tracking that we kind of mentioned before with the whole Facebook debacle, some transparency requirements, and I think also Google's kind of forthcoming changes to user tracking in Chrome might have influenced Caden's favorable position because they've raised, I think, a total of$24 million to date.

10:02And Roa plans to expand Caden's New York team further and develop Caden AI and establish more collaborations with retailers and other platforms. I think that's what they're really spending the money on. And this is what he said specifically about the value proposition, quote, we offer one of the most sought after assets in the enterprise ethically sourced first party data. So I think while marketers might be drawn to platforms like Caden, I think it remains to be seen if users are going to follow suit, right? Obviously, this is very valuable for marketers and AI platforms. So it's going to be interesting to see how many users are actually able to attract, how lucrative that offer of payment for data is, and if they're able to really grow this and expand this in a viable way.