The One Question bounty closes Thursday 10 December.

Join in
We Them Media logo We Them Media Support us

The Future of Privacy - Olayinka Oshidipe

The picture loads from YouTube. The player loads only when you press Play. Turn on captions in the player if you need them.

Olayinka Oshidigbe, who works in developer relations at Nilion, argues at Web3 Lagos that people trade privacy for convenience and that apps are built to extract data. The talk outlines privacy tools such as zero-knowledge proofs and fully homomorphic encryption, describes a private chat product, and urges developers to start from user privacy.

Not by We Them Media Channel: Web3Bridge. Published Monday 8 September 2025.

Titles are the owner’s. It plays from their channel.

Transcript

This transcript may contain errors. This video belongs to Web3Bridge. Letters mark each change of speaker.

These are the speakers’ own words. We have not checked their claims.

0:00 A: Olayinka Oshidigbe to speak to us on the future of privacy. Please, a round of applause for him.

0:06 B: All right. Hi. Good morning, everyone. My name is Olayinka. Um, the title was supposed to be the future of privacy, but I changed it last minute to be the future of your data because, like, it's one and the same, right? Um, so quick intro about me. My name is Olayinka. I am DevRel, Developer Relations at Nilion. At Neilion, we are ensuring that your data stays private. I'll get into that in a bit. Does this work? Oh, this thing is so cool. Okay, so we've been taught to trade privacy for convenience. What does that mean? That means if you've ever used any social media platform- Instagram, TikTok, Twitter, any of those things- you know that They are very convenient in the sense where, especially Instagram, where you get ads for things you're looking for, but like they are selling your data. So I personally tend to buy a lot of things from Instagram. I buy a lot of shoes, right? So if I'm shoe shopping and I go on Instagram, or maybe I'm having a conversation with a friend, the next thing I know I'm seeing a bunch of ads on shoes. Will I buy them? Absolutely. But that's not the point. The point is that my privacy has been violated, right? They figured out I was looking for shoes because of the conversation I was having with a friend. While it is convenient, right, I'm trading my privacy for it, and that doesn't need to be the case. And so at Nilium, we are building the technology that ensures that you do not give up your privacy to have the experience of convenience. Does that make any sense? Right? Okay. So, now the problem is all of this is by design. Like all of the applications that are being built today, they are built with extracting your data by design. It's not an afterthought. When Instagram, you know, was built or started up, whatever, they needed your information, your email, password, yada, yada, yada. And also, Meta, Facebook gets audio data, if you will, from your smartphone. If you think I'm kidding, just hold your phone. You don't need to do anything. Hold your phone and say things like restaurants near me, you know, or food, rice, whatever. And in a couple minutes, you'll begin to see those ads pop up on Instagram. How? They are extracting your data. They are violating your privacy. And then they are selling those things to advertisers, and that's how they're able to serve you ads, right? So privacy isn't missing, it is intentionally excluded. Now, a lot of the things I've talked about, I've talked about like Instagram, Twitter, um, and the likes, but we have a more bigger problem on our hands, and that is the problem of chatbots, AI, ChatGPT, and everything in between. In the past couple months, a lot of really crazy things have happened. One of them is hundreds of thousands of, um, Grok chatbot conversations are showing up on Google search. Here's what that means. That means that if you put in a query into ChatGPT or Grok or whichever one of these AI chatbots about your personal issues, it can show up on Google, which means I can see what the problem you're facing. And you don't want that. Like, you really do not want that. And it goes beyond, you know, your personal issues. It goes into your medical records. It goes into whatever sensitive- anything you consider sensitive that you share with any of these chatbots is very- it's liable to be exposed, right? And again, we're coming back to the, to the point of trading privacy for convenience. Now, I use ChatGPT. I think, you know, most of us use that, and it is a really, you know, efficient tool. It makes your life ever so slightly easier. But again, we're trading all of that convenience for our privacy. And as you can see, this one is actually really huge. It's like, um, some guy, a crypto millionaire in Europe, was attacked and almost lost everything, and it cost them just $6,000 to get the data they needed to kidnap him. And he's worth millions, so $6,000 is a walk in the park. But, and then there are some of us that are obviously not worth that much, that they can probably get our data by just running a Google search, right? So your data is being sold, and there is a solution for that. You do not need to live in a world where you're trading, constantly trading your privacy for convenience. And, um, the first speaker, when he was talking, he talked about, um, chatbots and privacy and everything in between. And it is super important that as we are exposed to these technologies, we also understand the underlying issues that comes with them, right? So now, while we have The current issue of privacy and our data being sold and all of these companies basically just ripping us off and making a lot of money from us, we also do have privacy-enhancing technologies that solve this problem. I'll give us a very quick high-level overview just so we know what we're talking about. So we have zero-knowledge proofs. Is anyone here familiar with ZKP, zero-knowledge proofs? No. OK. So zero-knowledge proofs are a way to prove that you know something without revealing the data. I'd encourage you to do a deep dive. You can look it up. It's very simple and straightforward. But I won't be doing that in this talk. I've done it a couple of times prior. We also do have trusted execution environments. A trusted execution environment is an environment where secure computation happens. We have multi-party computation. No one holds the full data. only partial shares. So basically, your data is split across multiple nodes. Then we have my favorite and it stops working. Did I break it? No. We have my favorite, which is a fully homomorphic encryption. Can you please help me move it to the next slide? I know. Thank you. Fully homomorphic encryption computes encrypted data and it doesn't need to be decrypted. I'll just spend a couple minutes on the last one so I can explain to you why it is really cool. So take Instagram, for example. Currently, Instagram needs everything from you to process anything. And that means even the engineers are exposed to your data, right? That means the government could call Instagram today and they're like, you know what, we need everything this user has uploaded to your servers from day one. And they will provide it. Easily because they have it. Do you understand? Now, some of us have deleted pictures way back in 2010 when we had questionable looks. Um, we were in very different points in time, and those pictures should stay buried. Instagram will bring them up, I promise you. Same with Facebook and every other platform you've used. Now, a fully homomorphic encryption is a piece of technology where you send in encrypted data and that data stays encrypted and it is processed and you get a result. Basically, if I hand you- if I hand you a recipe to cook food, you don't need to look into the recipe to see what it is made out of. You are able to make the meal without knowing what the recipe is, right? So that is a high-level overview of what a fully homomorphic encryption can do. And it is really cool because it means that any- in between now and the next couple years, we have applications that are taking advantage of fully homomorphic encryption, and we don't need to sacrifice our privacy for convenience. Now, one of the really interesting tools we built recently at Nilion is a tool called It's not working. Please, next slide. It's a tool called NeoGPT. Okay, before we get into that, we need to build products where privacy is the starting point and it is not a last-minute patch. So if you are a developer here or an enthusiast, if you are looking to build stuff, and, and, and I think I encourage you to, you know, begin looking to building infrastructures, begin to look into building technologies, because that is one of the many ways we can get Africa Nigeria on the map, right? Anyway, if you are doing any of such, I would like to implore you to be sure that you are thinking of the user privacy. What that means is you're building an infrastructure that doesn't expose your users to threats. That doesn't expose- I mean, I just showed you the thing where they basically kidnapped a guy. You know, and they were exploiting him of his money. And the onus is on obviously the developers to build systems and infrastructure and products that are privacy-centric. Um, and like I was saying, one of the- next slide, please- one of the tools we're building over at Neon is called NeoGPT. So NeoGPT is basically a very privacy-preserving version of ChatGPT, right? So all of the things I said about ChatGPT were You put in queries and Sam Altman can practically read your conversations. None of that happens with this piece of software. And this is possible because we are building a technology called NeelDB and NeelAI, right? So, so we, we have- we've built the infrastructure that ensures that you're able to build user products that covers your user privacy. So the technology, the engineering team behind, um, the engineering team at Neleon that is building this, we don't see any version of your data. If you can move to the next slide. Does it work now? Thank you. All right, sweet. Uh, It's taking me backwards. Oh, and there we go. And now it's frozen. Okay. Wow, this is so much fun. All right. It's broken again. Nah. Anyway, I'm just going to go on and speak to you and hope that this actually works. Okay, so the next- please stay at slide 7. I think it's slide 7 that I'm talking about. So, NeoGPT. If you've ever used ChatGPT, can I see, you know, show of hands? I've used ChatGPT at one point in time. Okay, that's a lot of us. That's everyone. Exactly. Now, what this means is every single one of us, at least 90% of us, have exposed our data to the OpenAI team. This should terrify you. I'm not even trying to be dystopian or whatever. You should be scared. Now, you can say, well, I mean, who am I? I'm just some random Joe. There isn't anything about me that is spectacular. 5 years from now, would that still be the case? No, because you'll be far further in life than you are right now, and you want to ensure that you have a track record that no one can take advantage of, right? So NeoGPT is your totally secure AI companion. And again, I'm not really coming here to say this, oh, I'm trying to get you to use NeoGPT because like I work here. Honestly, it doesn't really matter. You don't have to use it. However, This is a ridiculously cool piece of technology that ensures that everything you are saying and doing stays private. You get the best of both worlds, which is privacy and convenience. And if that doesn't trigger something in you, then we do have a long way to go as far as building products that everyone gets to use while ensuring that it stays private and secure. Right. From user verification to- we also have ensuring you have passphrases. Passphrases mean if you forget your passphrase, you don't have access to your conversations. And the backend team don't even have any idea what is going on in the frontend. Because like I said, a bunch of technologies, privacy-enhancing technologies that we are making use of. Fully homomorphic encryption means all of that processing is done encrypted. It's done in an encrypted environment. And no one, nobody, no one sees anything you are doing. And so this is just one of the many things you can build. You can build an expense tracker, you can build, um, a medical record where if you're in the medical field or if you have a friend in the medical field where you build a product where you're passing in user data and it's not being exposed to people who shouldn't see them. If you've ever worked in a field where data is sensitive, where data is crucial, where secrecy is important you understand the importance of isolating certain parts of data and keeping it from people's eyes, right? And so ensuring that you're building products that keeps your- considers your user privacy is really the next frontier. And I want to encourage us to begin to think about what that means as you, you know, are exploring the applications you use, as you are building or thinking of building products. you are looking at your user privacy. So finally, a privacy-first future starts with us. If you can put it in the next slide or not. Yeah. I also encourage you that as you leave here, my talk is not really technical. I had no intention of making it technical. I just wanted to inform you of what's happening. Your data is being sold every single day. Your privacy is being violated every single day. And you need to take ownership of that. And one of the many ways to do that is to use applications that are privacy-preserving and to take control of your data and your privacy at all times. So thank you very much. I hope this was helpful. My name is Yinka.

More from this hub

Something wrong, or a video you want off this page? Write to wethem.xyz@gmail.com and it is fixed or taken down within 48 hours.