Profiting from Privacy

Transcript

Hi. I can barely see you all because of the lights, but um yeah, good afternoon. I am Liz Dyninger from Least Authority. For the past 13 years, I've been talking to people about why privacy matters and how to build systems with privacy by design along with how important it is for our systems to be secured through the use of transparency with open source code and independent published verifications of security posture of these products. But today, I want to talk to you about profiting from privacy. As the CEO of a for-profit company for almost 10 years, I've been thinking a lot about making money.

>> [laughter] >> So, um I believe that the time is now for us to take our noble sense of privacy as a human human right and embed it into profit-making approaches. We need to deploy products that are private by default as the new economic model. I expect that you were if you were drawn to this talk, it is because you are all familiar with surveillance capitalism and would like to see change. Well, lucky for us, the days of data brokerage and and shitification are numbered.

Both thanks to the increasing awareness of the issues and the imminent threats of AI and agents. Let's quickly review how the popular business model that is driving the current web is doing. Business centered on the collection, analysis, and monetization of our personal user data user user data has seen impressive growth since the early 2000s. However, the response has been increasingly harsh. From the GDPR to the recent US court cases where social media companies were found to have harmed mental health.

The Cypherpunks created technology that aimed to respect privacy and crypto promised an alternative with web 3 infrastructure. With web 3 trying to provide a new infrastructure. However, privacy-enhancing solutions are only now getting more attention and discussion. Since 2016, 10 years ago, um when Zcash implemented zero knowledge proofs in production, we've had the knowledge of how to utilize the advanced cryptographic building blocks like this.

Yet, we are still seeing many mainstream products mimic the issues found in web 2. Surveillance and a lack of concern for user data has become the default as it was before. To move beyond this, we need to have a more nuanced conversation about what really should be transparent and what should be private by default. I find it helpful to think about the power dynamics holding these those with power accountable while providing agency to those previously without it. To oversimplify it for the sake of brevity, we need to make transparent how the system functions while the privacy control of personal data remains with the user.

The Cypherpunks manifesto captured what we had known for more than a century and has been reignited by surveillance capitalism. Privacy is not an individual affair for the one who has something to hide. It is a collective endeavor and like all human rights, our society suffer when it is not upheld by our day-to-day interactions. The opportunity for us is selling these three points, which can sell send seed new business models for the next era of systems.

And again, for the sake of time, I will only provide an introduction to these and hope that more discussion of the nuances in the future takes takes place. We can sell our products using credibility we earn by being fundamentally committed to our value of user sovereignty. We can reduce costs by maintaining compliance and minimizing the attack surface of our products. I imagine you've heard this positioning before. Privacy, the real private real privacy is something that users are ready and willing to pay for. Beyond that, if we can show users that the value that we value their privacy, their loyalty will result in lower costs for acquisition.

And of course, the risk of attacks will be reduced and the ability to comply with privacy regulations will be increased if a system is built with privacy in the design. Legal and policy frameworks can create obligations, but system architecture ultimately determines what can be enforced and verified. Signal and Tor are good examples of this. However, they are not not-for-profit companies. Although their choice to be not nonprofit makes sense, we can expect cannot expect that the future of our technology can be adequately competitive without profit.

And there are numerous web 2 products that have done a decent job of addressing these selling points. In the case of web 3, the practices of open source code, published reports, and new governance models are critical for earning credibility. And security efforts have improved over the years with many projects getting regular security audits and decentralizing the control. However, the use of privacy-by-default approaches is still lacking, resulting in unfulfilled promises about privacy being a key feature of the web 3 revolution. And although data protection rights aim to empower individuals while restricting how organizations use and process data, in practice, the experience of control often falls short of expectations.

The organizational and accountability-based model of control embedded in regulations like GDPR does not fully reflect the operational logic of modern day data systems. The tension becomes particularly visible and complex in highly distributed processing environments. In web 3 systems, data is routinely replicated across services, jurisdictions, and vendors, complicating efforts to maintain comprehensive oversight of access and use. Questions of control are increasingly shaped by how data processing systems are configured to manage the data.

There are certain technical measures which show that access restrictions can in many contexts be enforced directly through system design rather than relying exclusively on legal obligations or subsequent oversight. By utilizing these approaches, we can develop more robust types of revenue streams from limiting the data shared and ensuring unlikability of the data that is shared. We can reduce or even remove the need to rely on complex legal contracts and the challenge of enforcement in multiple jurisdictions.

There are a variety of technical measures that can be utilized to implement new data management approaches such as end-to-end encryption or user-managed credentials along with other cryptographic techniques including zero knowledge proofs. However, it is the deliberate and selective combination of these measures that can prove that can create new paradigms for business opportunities. For example, we created Double Blind Armadillo for Freely, which is a new wireless service that wants to know as little about its users as possible. Double Blind Armadillo is a new privacy-focused system architecture that and cryptographic protocol designed around the principle that no single party should be able to connect an individual's real identity,

payments, and phone records. Customers should be able to access services, manage payments, and make calls without having their activity tracked across systems. It works by partitioning critical information relating to individual customer identities, payments, and phone usage into separate service components that only communicate through carefully controlled channels. Each part of the system only knows knows only the essential information that is required to perform its job and nothing more.

This approach enables unlikability, which ensures that even different parts of the system even when different parts of the system must cooperate to provide a service, no single party can piece together a full picture of who the individual is and what they do. Unlikability thus prevents the creation of comprehensive behavioral profiles and in turn drastically reduces some of the risks commonly associated with data exposure. This is just one example of how the business functions can shift as a resulting as a result of implementing the privacy property of unlikability.

So, to recap, we're at the end of an era of surveillance capitalism. The fragile systems that we that were created are not sustainable for the long term and the adversarial nature of the relationship between the company and the user has resulted in the enshitification of the products we rely on. Besides selling our credible commitment to privacy and the ultimate savings from risk reduction, we can also create new types of revenue generation and new market opportunities when we design data supply chains to function differently.

We have an immediate opportunity. The default does not exist yet. We are no longer directly competing with the incumbents as the landscape is rapidly changing. Implementing data supply chains that prioritize privacy privacy can allow us to profit in new ways. I could end here, but there is a bit more that I would like to add. So far, I've not mentioned the elephant in the room of the world, AI. [laughter] When with the expanding use of AI agents, there is a another concern that grows. Besides the confidentiality of our data, we are also realizing the integrity of our data matters more than

ever. If we re-architect our relationship with data management and infrastructure and products, we may be able to address the issues being at presented by the ever-growing use of AI agents. The principle behind Least Authority's name itself, the security concept of least privilege, applied as a design philosophy. Don't give power that isn't required. Don't hold what you don't need. Don't expose what you don't have to. So, I hope that together we can launch new products that profit from privacy as a way to rebuild the digital world so that in that so that business

incentives are better aligned with our fundamental human needs. This is my contact info. And I'm happy to answer any questions. And thank you all for attending and thank you to the ETH CC team. Does anyone want to talk about this? Any questions?

If not, then thank you. >> [applause]