Speaker 0
0:10 – 0:14
Welcome to Tech Talk. Bye. CT. Tea.
Speaker 2
0:17 – 1:10
Welcome to CDT's tech talk where we dish on tech and Internet policy while also explaining what these policies mean to our daily lives. I'm Brian Wasilowski, and it's time to talk tech. Did you miss CDT's Tech Prom? Well, fear not. In this episode of Tech Talk, we bring it to you. Our president and CEO Nuala O'Connor gave remarks at Tech Prom where she challenged everyone to think about how they could help create technology that advances democratic values. Nuala's inspiring words are the first segment and worth a relisten even if you were in attendance. After Nuala, we'll be talking about a a new network of researchers, scientists, and academics who will be working directly with policymakers to infuse AI policy proposals with more actual research. Let's get to it.
Speaker 0
1:14 – 13:39
You all so much. You know, being in this big concert hall kinda reminds me back in my acapella days in college. Yeah. No. I'm not gonna sing. That's what White Ford Bronco's here for very shortly, but we here at the Center for Democracy and Technology want to thank you. Thank you so much for your support of this great organization over the last twenty five years. Sure, we can throw a really great party, right? But even more, we know how to make real change happen, and we do that through the slightly less glamorous work showing up every day in meetings and negotiations, in conference rooms in Washington and Brussels and Silicon Valley, in hearing rooms and courtrooms around the world, the Center for Democracy and Technology fights for the rights of the individual in the digital age. This work matters more than ever. Democracy and democratic values are under attack around the world, and while technology increasingly informs almost every aspect of our daily lives, democratic values are not yet fully embedded in every aspect of technology. We believe that technology can and must inform and advance democracy. This is the heart of CDT's advocacy here in The United States and in Europe, where just this year, CDT was formally incorporated as a recognized European NGO, CDT Europe. We believe technology can serve the first principles of democracy, and so we are working to make elections more secure around the world. We believe technology can advance the individual rights that are essential to a democracy, and so we push back against calls to weaken encryption. We believe that technology can support an informed and an educated electorate when we call for stronger privacy protections for student data and better technology in the classroom. And we believe that technology can respect and advance human dignity when we challenge warrantless searches of devices at the border. Fundamental to a free and open and thriving democracy is the right to individual privacy. Those of you who were with us last year at this dinner will remember that we called for comprehensive federal privacy legislation in this country, and at that time, it probably seemed to some of you at least, maybe a little bit optimistic. But look at where we are now. Someone one person clapped. Yay. Consumers are aware. They are concerned. States are taking action, and Europe has already acted. The notice and choice model, it is dead, my friends. The United States needs fair and clear rules across industry sectors, Yes, that allow innovation to flourish, but more importantly, respect the dignity of the individual. The Center for Democracy and Technology was one of the first movers in this space back in 1994 when it was founded and again this year as we brought together all of our stakeholders to draft model legislation that has already moved and shaped the debate. Congress needs to act now. We will get this done. Freedom of expression and information are also fundamental to the democracy, and CDT works every day to promote the Internet's best and highest use as a platform for individual speech. We work every day to combat censorship of all kinds and to address the real challenges that online speech creates, and we must all confront together the scope, the scale, the shape of misinformation, extremist content, and hate speech. And there are no easy answers to this one, my friends, which actually makes it the perfect issue for the Center for Democracy and Technology with our deep domain research and our academic partnerships and our working groups and most of all, our ability to listen and to learn from people on all sides of the conversation. As always, as you see, the Center for Democracy and Technology team is hard at work, and there is yet so much more we can all do together because it is certainly true that technology can be an equalizer. From voice activated devices assisting people with mobility or visual impairments to connected classrooms to data enhanced health decisions, technology is helping improve lives and, yes, democratize society. In my life, ubiquitous technology allows me the ability to both work and to parent, to fill two full time roles, not always with the grace that I would like, but on many days to be virtually present in two places at once. Technology is an equalizer of opportunity for me, and it is for so many others in so many different ways, but it is time. It is time to move beyond this narrative that technology is an unfettered good and move into a discussion of impact. It is time to talk about power, the power of data, the power of platforms, the power of scale and network effect. It is time to talk about the harm that our digital communities are having on real people in the real world. For too long, our fundamental assumption has been that the Internet's open design is inherently equal and thus would make everything better for everyone. It is time that we acknowledged that this has been a limited and somewhat privileged view. It's time to move beyond this performative equity equality into the harder conversation around equity because equal does not always mean equitable. Inherent in the very architecture and infrastructure and design of many parts of the internet are the values and experiences and beliefs of the creators. We all create things. Some of us create art. Some of us create music. Some of us create children that reflect us and often look like us and hopefully inspire us. But not everyone can see their reflection in this thing that we have increasingly embedded opaquely in so many people's lives. Our digital world is architected to work for some people, perhaps a certain race or a certain class or gender or nationality and that's super great if you happen to be one of them. But in a democracy, the institutions, including our digital ones, are supposed to work for everyone. Equity is essential to a thriving democracy. It is time that we each embrace a greater individual and institutional responsibility in our digital age. Each of us has a duty of care to this democracy. So here's our challenge for you. How are you, yourself personally in your work or your institution, how are you informing and advancing the democracy? Policymakers, and I know there are many of you in the room from The United States, from Europe, Commissioner Yerova, and others, wherever you sit, are you challenging your assumption about how real people in the real world are actually using the technology? Are you thinking through the unintended consequences of your decisions? Are you considering the real world impact of the online spaces we are creating on all of the beautiful online space, a diversity of communities? And whether your product is an online game or a job search platform or something else entirely, are your creations informed by values of equity? And personally, what percentage of your time are you spending creating products that are solving the world's hardest problems? Some of the attributes of our online spaces have been volume and velocity and scale, and while louder, faster, and larger works really well for concert halls like this, these are not the only attributes that we should consider when we are fostering a healthy democracy. We must work to embed the fundamentals of democracy in our institutions and our communities, including our digital spaces. We need to design for democracy. It's on us all of us to do better, and when it comes to the work of sustaining democracy, as the saying goes, none of us are obligated to complete the work, but nor are we free to abandon it. More than twenty five years on from the dawn of the commercial Internet, the dialogue around the technology in our daily lives has profoundly changed, and CDT has been here for all of these past twenty five years, and we are looking forward to the next twenty five to help you craft policies and laws and design technologies and build norms and institutions that will advance democracy. Because as the saying goes, this is not the end. It is not even the beginning of the end for technology, but it is the end of the beginning. This thing is built. Now let's make it better. Thank you.
Speaker 2
13:46 – 14:39
AI seems to be the tech buzzword du jour, with many policymakers either super excited about the potential to save humanity or terrified that it will lead to robots ruling the world. What seems to be lacking in many of the conversations around AI and policy, though, is actual research. CDT and the R Street Institute are working to change that through their newly formed GRAIL network, which stands for governments governance and research for artificial intelligence leadership. You can see why I stumbled over that one. The network will facilitate engagement between policymakers and the scientists, researchers, and academics who are working directly on artificial intelligence. CDT's GRAIL lead, Stan Adams. That's a great title. You want that one, Stan? I'll take it. Alright. Well, he's here on Tech Talk. It's a good title. Let's talk about the GRAIL network. Top level, why did CDT and R Street come together to create this
Speaker 1
14:40 – 15:29
thing? Great question. So both CDT and Art Street think it's important for policy to have a basis in the realities of technology, in terms of how it works, what it does, what it can and can't do. But staying current on tech, especially something as complex and evolving as AI, is tough for even those of us who work in it. So we wanted to help policy makers by providing a resource that they can turn to, when they need up to date information on the state of the art of AI technologies, and a way for them to get in touch with sort of the leading thinkers in the space. So together CDT and R Street represent a bipartisan team, but the network is actually nonpartisan, meaning that it does not espouse any particular political viewpoint, but instead provides the best information available.
Speaker 2
15:30 – 15:49
It is up to the policy folks to decide which way to lean. Oh, that's good. I like that bipartisan nature. Well done there, our grail lead. So buzzword, I mentioned early on people throw AI out there all the time. What are you all talking about when you use the term AI? Do you have a working definition for your work?
Speaker 1
15:49 – 16:10
So I personally don't think it's all that helpful to have a hard definition of AI. For the purposes of the network, AI is anything that policymakers think it is. Oh, wow. We we want to be able to help them answer questions even if they're misguided or not exactly and what other people think is AI.
Speaker 2
16:12 – 16:42
Basically if they have questions, we want to try to answer them. Cool. What do you think is like the most ridiculous definition of AI? You don't have to answer that. Definitions of AI? I always think people go to robots ruling the world and I think that's just one that I'm like, well, I don't think it's gonna be Terminator level yet so, that seems a ways off. Yeah. But maybe I'm the first one killed if that happens. So, so you mentioned this is a network of academics. Who are some of the the
Speaker 1
16:52 – 18:03
academics who work on AI. Cool. Name drop. They are, University of Washington. Christian Lum is a leading data scientist working on fairness, accountability and transparency in AI. She's especially focused on systems that impact human rights such as predictive policing and pretrial risk assessments. We have Robert Siemens of the New York University School, Stern School of Business. He's an expert thinking about the impacts of AI on the economy. We have Suresh Venkatesubramanian, computer science professor at University of Utah. He works at the intersection of machine learning and fairness. We have Margaret Hu. She is at Washington and Lee University. She works at the intersection of data and national security and surveillance, and and and their impacts on human rights. And we have James Besson of Boston University. He works on innovation, automation, and thinks about those things from an economic perspective. Wow. That is quite the group. Impressive.
Speaker 2
18:04 – 18:16
So, I mean, obviously, in describing them a bit, you you touched on this. But what are some of the big questions that these researchers are trying to answer? Ones that, you know, you're particularly drawn to. Sure. So so right now, what
Speaker 1
18:17 – 19:23
the government here in Washington DC is focused on is sort of how to incorporate AI and other automated systems into the government now. And so the bigger questions revolve around the role of AI in government such as how and where the government can or should use AI systems, including issues to consider before they buy these systems. Right. Questions they should ask of the vendors that they may want to think about before they before they purchase. We are also thinking about, or anticipate working on how the government should think about standardization in the AI context. The National Institute for Standards and Technology, NIST is currently, taking public comments on standardization in the AI world. This is part of the president's executive order from earlier this year. And then we will also be turning to how to promote access to government data sets,
Speaker 2
19:24 – 19:31
for machine learning. That's cool. Does ethics play a role in any of what they're researching? Of course. How so?
Speaker 1
19:32 – 19:35
Ethics is maybe the number two buzzword behind AI.
Speaker 2
19:37 – 19:42
Sorry I brought it up. So it's We're not sorry. Like everything we work on, it's complicated. Yeah. Right.
Speaker 1
19:45 – 20:12
Right now, ethics are largely being defined by the companies that build the systems. And so we are in a little bit of a space where we have to take their word for it. Gotcha. And so we would like to move beyond that and have more, outside views on what is and isn't ethical. We would also like to try to achieve some sort of consensus on what that is Yeah. Moving forward. Sounds like why a multidisciplinary
Speaker 2
20:13 – 20:33
approach to AI and policy making is needed. So GRAIL network, here it comes. Comes. How in terms of operations, how are you going to directly connect these brilliant researchers and academics to policy makers? What are kind of the interventions you're gonna gonna create? Sure. We we hope to do a few things.
Speaker 1
20:35 – 21:24
One, we hope to bring them physically or virtually, here to DC and elsewhere where policy making discussions are happening. Like Brussels where CDT has an office. Exactly. And Europe is definitely thinking hard about what to do with AI moving forward. We also help to, we hope to help researchers engage in policy discussions on their own, right? So we want to give them tools and resources that they can use to to engage in an otherwise potentially unfamiliar process. Things that we do here every day are maybe not what they work on in their day to day. And then we also hope to translate the work that they do into more accessible formats, for policy makers. So reducing a 50 page, academic study into
Speaker 2
21:25 – 21:51
maybe a page or two of the high level points and what it means for I mean, that would be hugely valuable. If you do that, I'd like to read that too because those academic papers can be slogged sometimes. Absolutely. I'm glad they exist, but, oof, they can be tough. So if there are people listening who are academics or policy makers who wanna tap into this network or academics who wanna be part of it, how should they get in touch with us? How can they become involved? Send us an email.
Speaker 1
21:51 – 22:00
You can email the network@infoatgrailnetwork.org, or you can email me stancdt
Speaker 2
22:00 – 22:35
dot org. I mean that is a great to hear from you. Awesome. Well, bravo. This is great work. Glad that it's a bipartisan one too and certainly as this network grows we'll be sure to have some of your brilliant researchers on and share their work excellent thanks for joining thank you That's it for this episode of Tech Talk. You can watch Nuala's Tech Prom remarks on CDT's YouTube channel if you want more of them. And you can also find out more about CDT's work on AI at cdt.org. I'm Brian Wasilowski. Thanks so much for listening.