Talking Tech on Trust, Power, and the Future of Technology Governance
What does it take to build an open, inclusive, and trustworthy internet in a world where technological power is increasingly concentrated in the hands of a few?<br>In this episode of Tech Talks, CDT President and CEO Alexandra Givens sits down with Sally Wentworth, President and CEO of the Internet Society, to explore the challenges facing the open internet and the future of technology governance. They discuss growing threats to internet fragmentation, the rise of digital sovereignty and online walled gardens, and the challenge of protecting users’ safety without sacrificing privacy, anonymity, or encryption. They also examine emerging cybersecurity threats, the impact of AI, and the importance of defending the multistakeholder model that brings together governments, industry, civil society, and technologists to shape the internet’s future.<br>As trust declines and technological change accelerates, can collaboration still deliver meaningful, inclusive decision-making? And what will it take to ensure that everyone — not just those with the most power and resources — has a voice in shaping the technologies that affect our lives?
Top Keywords
- internet 0.014
- internet society 0.010
- safety 0.008
- encryption 0.007
- isoc 0.007
- work 0.006
- society 0.005
- safe 0.005
- online 0.005
- keep safe 0.004
- around 0.004
- model 0.004
Transcript
Welcome to CDT's Tech Talks, where we dish on tech and Internet policy while also explaining what these policies mean for our daily lives. I'm Jamal Magby, and it's time to talk tech.
Welcome to Tech Talk by
CT. Team. Today, we're talking about trust, power, and who gets a say in shaping the future of technology. The collaborative multi stakeholder model that helped build the open web is currently under strain even as technology becomes more powerful and decisions are increasingly concentrated in the hands of few. For this conversation, I'm handing things over to Alexandra Givens, CDT's president and CEO, and Sally Wentworth, president and CEO of the Internet Society. Together, they'll explore what the web's collaborative past can teach us about the future and whether we can build a technology governance model that is more inclusive, legitimate, and worthy of public trust. Can collaboration still work in a world where trust is low and power is increasingly concentrated?
And what will it take to build something better? Let's get into it. Before we start, I'd love for you to talk a little bit about the Internet Society, what it does, and why it matters so much for Internet users all around the world.
Sure. So the Internet Society and the Internet Society Foundation, we are a global mission driven organization that has, a couple primary, goals. The first is that, we want to see everybody connected to the Internet, to an open, secure, trustworthy Internet for everyone. And so we worked very hard to ensure that people have access to affordable, reliable, and resilient Internet infrastructure. We do that through advocacy for Internet policies that, support an open global Internet. We help communities build the infrastructure they need for Internet access in some of the hardest to reach places on Earth. We support the development of digital skills to help people use the technology safely. And then finally, a a a fun fact about the Internet Society is that we are a global community. It's sort of what sets us apart and makes us unique. We have over a 110 independent chapters of volunteers who are passionate about the Internet, who work every day to advance these goals, and over a 150,000 members who join us because they believe in this mission and in this purpose. And so all of that combined is really what shapes the work that the Internet Society and our foundation do. It's a big mission and, obviously, evergreen. What are some of the priorities that you're most focused on right now? So we have a 2030 strategy, that was adopted a couple years ago that really guides the work that we do because, as you say, you could be the everything society, and we have to stay focused. So our two key strategic priorities are, first, that people everywhere have access to affordable, reliable, and resilient Internet. And second, that people everywhere have an Internet experience that's safe, secure, and protects them online. And so those two strategic goals really guide all of our our programming, our philanthropic giving, our training and capacity building, and all of our community outreach around the world. The safe, secure
environment is is the area where CDT works with Internet society the most, and I'd love to drill down on kind of some of the challenges and changes that you're seeing in this space. This conversation also could boil the ocean. These issues are so lies in people's minds every single day, but let's let's dig into a couple of them. One I'd love to start with is fragmentation. I mean, one of the major changes since you started at ISOC is the growing threat of Internet fragmentation. I'm curious how you view the growing public policy conversations around data localization, digital sovereignty, these concerns that because of the direction in which the Internet is going, the answer needs to be kind of throwing up these additional walls and moats around people's data. How do you think about that and the Internet Society's traditional vision of an open interoperable
Yeah. It's a real big challenge that's facing us now. You know, for a long time, the discussion kind of was, are we in a fragmented Internet or are we not? And I think the the reality is that fragmentation is often cumulative. It's not something that just happens one day. We wake up and this has happened. It's really accumulation of policy developments or technical technical choices or commercial decisions that are driving us in a direction that we think is really the wrong direction. So things like routing restrictions, data localization mandates, service blocking, the sort of patchwork of laws that that are emerging around the world that really work against this idea that the Internet was built to be decentralized. It was built to be a global network of networks where networks voluntarily connect to this infrastructure and that there's value derived from that. And so, you know, it's a... It... It's always been a challenge, I guess, I would say, you know, yes, we're seeing it really front and center right now. But as long as I've been working in this space, we have seen governments really try to grapple with a technology that fundamentally is global and wasn't designed around national borders the way traditional telecom networks were. And some of what we're seeing, I think right now that's actually... I mean, it's interesting. Sometimes it's a little scary, but is government's grappling with their... I think what's kind of a natural desire to understand their risk profiles, their dependencies on a small number of of services or service providers, a desire to see more competition for service provision, the desire to have more agency over over their data, and I I think that's kind of natural and understandable. When that tips over into hard restrictions that go deep into the Internet infrastructure like we've seen with data localization mandates, And we... That's where we get really, really concerned because that's really where you start shutting off user choice. That's where you start choking off innovation, and that's where you start hardening the infrastructure in a way that is less resilient, less robust, less able to respond to the security challenges that are emerging. And so, you know, this isn't, again, a black or white issue. We're on a continuum.
But, certainly, I think the discussion is accelerated in a way that, you know, I... I've never seen before in my my history of working on these issues. There's a related threat emerging, which is maybe different in kind but similar in result in response to AI. We've been very focused at CDT on the growing use of paywalls and other measures to counter AI scraping. And much as you were just describing now, there is a very understandable reason why people are doing this. Right? The weight of scraping traffic, concerns about intellectual property. I mean, all of these different considerations as there is an increasing number of AI related crawling happening on the web. But we've also worried about the result being a greater reliance on walled gardens and it being harder than ever for users to be able to access information in the way that we historically have been able to do so across the web. I'm curious if you're engaging on this at all or how you think about it. We start from the premise that,
the Internet is at its best when it's decentralized and when people can choose the products and services that work for them without being locked in. I think this tendency towards walled gardens has been with us since the beginning of the Internet. There's always been this commercial tendency, this commercial desire to lock in a user base to control how that user base interacts with the technology. And then over time, that user base gets a bit unhappy and starts breaking out because innovation, you know, people desire new services. They want interoperability. And so we've seen that that tugged and pulled throughout the the course of the Internet. And so I don't say that to dismiss what we're seeing now, but it is something that that is a a, I think, a natural tendency that we've seen over time. The issue that you're speaking about right now, I think, is really important. And I I I guess, you know, ISOC directly doesn't engage on this, but where this is happening, of course, and a lot of the discussion around this is in the Internet engineering task force. And that's really important because the extent to which this conversation can happen in an open standards environment is a really good thing. So much of the dialogue around AI is happening within proprietary systems in these closed off environments. And so while this is a really, really hard problem, I do wanna at least acknowledge that it's happening in a space where organizations like CDT can participate actively, where publishers can participate, where owners of creative works can be part of the discussion. So I think that's actually a win. I think we should we should take that little win. I... I'd be interested from from your perspective at CDT is where you think that that discussion is going and how
how you're interacting in the in the process because I know CDT has been really active on this. I love that it's been five minutes and we have a plug for the standards bodies. I was expecting nothing left of you. There's there's some CDT listeners who will be denizens of the IETF and know all about this and others who aren't. And to them, I will say this, which is I will never forget the lesson I had one... Once I led what I thought had been a very successful advocacy meeting at the White House. And at the end of that, one of the advisers who was technically in power at the time came to me and said, Alex, administrations come and go, but standards are here to stay. And, essentially, what he was telling me was, doesn't matter how persuasive you just were, you need to be moving norms for the long term in the bodies where these are set. And that was a very important piece of career advice and validated the... This part of CDT's theory of change of... To your point, where are the technical experts that actually enable the architecture of the Internet spending time together and figuring out common and shared approaches in what is a relatively depoliticized... Has its own politics, but is a depoliticized environment. So on this scraping issue, the IETF has been having, a a work stream for a while now on exactly this question. And for those who are interested, it's the AI preference workings... Work stream. Built on this idea that people should be able to signal their preferences the same way historically they did via robots. Tech to decide what should be crawled for search purposes. I think the piece for us is figuring out how you inject enough nuance in that conversation knowing that it is not always gonna be black and white as to what people want their content to be used and scraped for and how you represent the perspective of the wide range of content creators for whom a paywall is just not gonna be the answer. Right? Wikipedia is not gonna throw their stuff up behind a paywall. They're not gonna want proof of personhood, I don't think, at least not administered in a traditional manner for people to be able to access that content because we want it Yes. To be ingested and we want it to be available. So where we've been pushing ourselves as a team is to think through, okay. Well, what are the other models that get appropriate recognition of the value of that human knowledge creation that, again, kind of restore agency to the knowledge creators and aren't just the existing tools in the toolkit that major publishing is always gonna push for, which is gonna be paywalls and and rigorous copyright protection. I don't think we have the answers yet, but to be honest with you, we're still actually building the field of participants in that conversation for this off the sidelines moment to say we need you in the conversation. So that feels like a really critical piece of the conversation ahead. I will say we also have this unusual moment right now where the EU is literally pointing to proceeding in the IETF to say, oh, this is hard. Can you guys figure it out for us? That really changes the dynamics of a voluntary standards body, to suddenly be, you know, potentially the vehicle through which legislation is being enacted by a backdoor. We've written about Kevin Banks from the CDT has written about that for folks that wanna nerd out and go deep on that... Some of those dynamics. But, again, I think it shows the importance of this issue, and it's not gonna go away as we really try and think through the future of knowledge consumption on the web in this AI era. Totally. But I think you... Yeah. A couple of things that you mentioned there are are really important. Everybody, of course, is talking about AI, and and there's a lot
of understandable fear and concern. But what you're talking about here is creating the conditions where the right people can participate in shaping the outcome. And I think that there's a lot of lessons to be drawn from that as we think about some of the other sort of thorny problems or challenges we're facing in the sort of AI industry. Making sure the voices that give people agency are part of the conversation. And I feel like that is often overlooked in the race to solutions is is really seeding the ground, as you said, and making sure that we have those right, stakeholders around the table in shaping a solution that really does give people agency over how their content is used.
I don't know if you intentionally teed up this question, but it makes me wanna talk to you about another trend or or two trends that we're seeing in kind of Internet governance with similar consequences. So one is the growing emphasis on age assurance as the solution to protect kids' safety online, and then another is proof of personhood to address various AI enabled threats. I'm curious how you think about this kind of balance of safety and security while still respecting respecting users' privacy and ability to access information? And who should be the voices in that conversation?
Right. You know, I'm a a mother of of of elementary school children, and so this question about online safety is one that is, you know, deeply personal actually to me as we try to navigate the space with with children. I think there's a couple things. We basically reject the idea that safety and privacy are in competition or in somehow intention with each other. We believe that secure and private communications actually keep people safe. And so putting that at the forefront of the conversation around safety is is absolutely critical. The way we've approached this at the Internet Society is that, you know, thinking about online safety design for young people or other vulnerable communities, We are a community led organization. That's really essential to how we work. We work in partnership with other organizations like CDT, to work through these kinds of challenges, and we work through consultations and dialogue with governments, industry, child safety experts around the world. We even convened leading safety and security experts to develop recommendations for keeping children safe and secure online. And we listen to parents. I have to listen to parents at ISOC because many of us are parents. And, and, you know, there's a couple things that we know. Obviously, there's this trend towards age checks in some form or another. Those can support safety tools, but they don't make a service safe. So we need to focus on, both safety of design, not just restricting access. And so as we were talking earlier about this kind of rush to an outcome, it seems like right now, I mean, there is this general public interest concern about child safety. That's understandable. I feel it. A lot of people feel it very deeply. But rushing to the solution of we're just going to restrict or we're gonna ban, those are kind of quote, unquote easy answers with a whole range of downsides. Right? These kind of solutions tend to require us to give unknown platforms huge amounts of personal data in order to validate our ages. And let's be clear that age checks does... Don't just apply to children. They will apply to every user. That's how you do it. And so, we want to ensure that the measures that are being rolled out, you know, maybe with very good intentions in mind, actually preserve privacy and security while addressing this this general public interest question. These are hard problems, and I think sometimes the rush to, you know, restrict as an outcome is, is misguided and may actually lead to to more harmful consequences down the line.
A natural follow-up there is to talk about ISOG's work explaining and defending the importance of encryption, where, again, sometimes we're confronted with this false dichotomy between privacy on the one hand and safety on the other. Yeah. CDT and ISOG partnered to cofound the Global Encryption Coalition several years ago. It now has over 500 members in countries around the world. We have our sixth annual Global Encryption Day coming up for all who celebrate. I'm curious why ISOC cares so much about encryption and how... There again, how do you explain that balance of trade offs when policy makers try to frame the encryption debate as this binary of privacy versus safety? So I... We completely agree with you on this. We just don't accept that privacy versus safety is the choice. We think that's a false choice.
And we understand, and some of this is part of our connections to, you know, the Internet technical community, that strong encryption is a crucial part of how we keep people safe online. I think a lot of people don't really understand or even see the ways in which encryption and encrypted services shape their online behaviors. But as more and more of our digital lives, our lives are being conducted over these digital services, we depend on encryption to ensure that the information we're sharing with different platforms and different providers is actually safe and secure and can't be broken. So my interaction with my doctor's office, my interaction with my child's school depends upon the the availability of strong encryption and the deployment... Wide deployment of strong encryption. So that is how we keep ourselves safe. That's how we protect speech. That's how we protect public services, critical infrastructure, journalists, human rights defenders, etcetera. This is a fundamentally fundamental technology that keeps us safe and protects our privacy. So, yeah, we've been at the forefront of this alongside, as you said, CDT and all of our partners in the Global Encryption Coalition because it is absolutely essential that this not get broken. And we're often confronted with policymakers that say, oh, you know, it's just for us. We, you know, we in the national security community, we just need this this access in. We just we just need our own special access point. They used to call it back doors. I think they realized that was a really bad term, so they they've evolved the conversation. But, fundamentally, that's still the thing that they're asking for. Right? They just want their special access in. And what we've seen time and time again is that a special access for one, it gets broken for everybody. And and so, we really have to double down on, you know, this is, this is fundamental for the digital economy. It's fundamental for online safety, for security. The availability of wide deployment of encryption is absolutely essential for a secure and trusted Internet.
On the safety front, last year, ISOC launched the Common Good Cyber Fund working to support NGOs whose work strengthens the public interest cybersecurity ecosystem. I'm curious about some of the threats you're most focused on there and kind of what the theory of change is.
This, really came, to us, from the Common Good Cyber Group. There was a group of organizations that were working together, to respond to this understanding that much of the work to keep us all safe online comes out of an ecosystem of nonprofits and volunteers, kind of the unsung heroes of the Internet in many ways, who maintain the essential services, that keep us safe. This is where they share threat intelligence. They're supporting incident response. They're securing core infrastructure and also working with high risk communities to withstand and survive online attacks. This is being done by a group of nonprofits that have traditionally been under resourced and stretched, far too thin. What the group also understood is that, obviously, cybersecurity threats affect all of us online, but we also know that they disproportionately target already vulnerable groups. So NGOs, journalists, human rights defenders, etcetera, are disproportionately, threatened by cyber cyber security attacks. So what the Common Good Cyber Fund does is, and when it came to us, it enabled us to put our resources in with major contributions from governments, philanthropists, and others to support this ecosystem of organizations that are out there maintaining critical cybersecurity infrastructure. We believe that these are absolutely essential for a trusted ecosystem for a strong and robust civic space that's needed in the multi stakeholder model. And so it's it's really important work, and we are enormously proud and honored to be a part of it. In 2025, we did a pilot round of funding, and distributed about, $1,900,000, to six inaugural grantees. And then this year, we're expecting to award about $3,500,000 in November to the next round of grantees. And, again, these are organizations that are supporting, some of the core cybersecurity
infrastructure that we all depend on whether we know it or not. How are you thinking about this work in light of AI enabled security threats, which have been obviously so front and center in the news cycle? You know, is that infrastructure layer ready to absorb this threat? And do you think the next few years will present more of the same, or is this really a step change in how we need to think about online security?
Yeah. I think, obviously, threats have always evolved, but what's different today seems to be around the the pace, the scale, the accessibility of increasingly sophisticated attacks. And so that's why we think this model that we're using here of pooling our resources really, really matters. It's important, and we've gotta make sure that the people doing this work that I described have the tools and the capacity to keep pace with these threats. This also includes making sure that these defenders can leverage AI themselves to identify and respond to the evolving, threat landscape. I'm not sure that the the issue really is the shortage of people thinking about these issues. There's a huge amount of experience and commitment, and we see that in the kinds of applications and grantees that are coming to us. The bigger challenge is that we gotta get the resources, into the hands of organizations that can turn that expertise into action and really make an impact. As you said, we're seeing that sort of step change, and we don't want these groups to feel like they're working in isolation or that they don't have sufficient resources to meet this moment,
because we're depending on them. I think one of the key elements in in what you're talking about there is this kind of vision for multistakeholderism. Right? Like, the role that industry has and governments have and this essential civil society public interest layer that can have durability and a different incentives over time. I'm curious how you're feeling about the overall model of multi stakeholders right now, you know, going from the original paradigm of the early web, which was built through this unusually collaborative model, bringing together technologists, companies, government, civil society, and users, to where we are now, which has required, like, some very rapid fire drills to stand up for this model of working together. I think about you in particular just because you've been so vocal in the effort to defend this approach, you know, speaking up in the run up to the UN's versus plus 20 review just to give one example, from the past year.
So I'm curious how you're feeling about the state of play. What does the work look like ahead to make the case for this mode of working? As as you said, the the Internet Society and many of our partners in the technical community, you know, this is this is... The multi stakeholder model is fundamental to how we work, how we develop technology, and how we support technology deployments around the world. It's not perfect. It has gone through changes. It's evolved, but I do think it gives us a framework of how to tackle hard problems. And, you know, I'm not one to hold it up as there's... You know, it's it's the model of perfection, but I do think what it does is offer us an understanding about how to bring different voices into the discussion. I will say, and I I said this, a couple of months ago in a in a piece I wrote following the cancellation of RightsCon. It's really important that, we in the multistakeholder models set up a notice when parts of the ecosystem are not able to effectively engage. And, you know, RightsCon is not the only way that civil society organizes and engages. That's absolutely the case, but it is one of the leading places where people meet in person to build that kind of public interest voice within the technical, ecosystem. And so it was, I think a moment to set up a notice that it is getting harder. The space for civic discourse is getting smaller, and it's getting harder for groups to meet. It's getting harder for them to communicate securely, and that that is going to be... That's to the detriment of the Internet and to the multi stakeholder model of Internet governance. So I... It's really important, and we at ISOC are really paying a lot of attention to that development and trying to do the part that... Play the part that we can, from the technical community perspective of saying, it is important that these voices are part of the Internet's development. It is easy, I think, to lose hope sometimes and to just get beaten down by, the press and by, you know, what we're reading about every day. At the same time, you know, one of the the beautiful things about the Internet Society is that we are a passionate community and many of of many volunteers who get up every day because they believe in the power of an open Internet. They believe that it can change people's lives for the better. And they're willing to to step up and and be heard on that on that topic, whether it's at a hyper local level, building technology, deploying online safety tools and training, or at the United Nations where we saw some of our chapters sitting on national delegations and saying this really matters to us. So I I don't give up hope, but I also am really, aware of the challenges and the pressures on the model today.
And I still think it's worth defending. I think that's a beautiful note to wrap on and to say, you know, CDT in our role, we've... We're still deeply involved in Internet governance and trying to take the lessons learned from the past and some of the newer conversations on AI governance. And I think it's voices like yours, organizations like ISOC standing up for this approach that we need structured spaces for all stakeholders to have a seat at the table and to do so with a grounding and technical understanding and with a deep commitment to users' rights. But there are important lessons to learn from that. So thank you, Sally, so much for your leadership. Thank you for your time. It's really been a pleasure to have you on the show today. It's been a pleasure having this conversation with you, Alex. And I just wanna say once again,
what an amazing team you have at CDT and just how much we value the partnership with organizations like yours and your leadership in in this space. It's it's really important that that organizations like ours, you know, continue to collaborate and work together because we have, common focus.
Thank you for listening to Tech Talks presented by the Center for Democracy and Technology. I've been your host, Jamal Magby. Tech Talks is edited by Jacob Kaufman and produced by Drew Courtney. Check out more of CDT's work by visiting us online at cdt.org and on various social media at sendem tech. That's c e n d e m tech. Thanks for talking tech.
