Smart Walls & Free Speech -- Talking Tech w/ Mana Azarmi & Emma Llansó
In this episode, host Brian Wesolowski sits down with Mana Azarmi to explain the civil liberties concerns that come with a proposed "smart wall" on the southern border of the US. Then, Emma Llansó shares with Brian the latest on legal challenges to FOSTA, the bill aimed at addressing online trafficking –– but that is having serious unintended impacts on free speech.<br><br>More on the smart wall from Mana: http://bit.ly/2SWX2fi<br>More on FOSTA from Emma: http://bit.ly/2SWXTN2<br><br>More on our host, Brian: bit.ly/cdtbrian<br><br>Attribution: sounds used from Psykophobia, Taira Komori, BenKoning, Zabuhailo, bloomypetal, guitarguy1985, bmusic92, and offthesky of freesound.org.
Top Keywords
- border 0.019
- plaintiffs 0.011
- speech 0.009
- border zone 0.008
- content 0.008
- fosta 0.008
- drones 0.008
- mana 0.006
- zone 0.006
- pinterest 0.005
- emma 0.005
- smart wall 0.005
Transcript
Welcome to Tech Talk. Bye. CT. Tea.
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. After a nice sabbatical, I am back. Thank you to our producer, Tim, for hosting, although I was definitely not working. He was amazing. Today, we have two of my favorite CD tiers as guests, Emma and Mana. We'll be talking with Mana about civil liberties concerns that come with a proposed smart wall on the southern border of The US. After that, Emma will share the latest on legal challenges to FOSTA, the bill aimed at addressing online trafficking, but that is having serious unintended impacts on speech. The president didn't get all the funding he asked for for a physical wall, but Congress did give him additional funding for a smart wall. CDT's policy counsel, Mana Ezerami, sees a lot of civil liberties issues with the blank check that congress handed over to Customs and Borders Protection. She joins us today. Hi, Brian. Now is this a big day for you, Mana? What is happening tonight?
Fleetwood Mac. What? Come on. Okay. Yes. I'm going to a Fleetwood Mac concert at Capital 1 Arena. Everyone should attend. And and you're a big fan? I am a big fan. Dating back to when you were just a wee youth? Yes. Since 1997, the dance counts a DVD. Check it out, kids. And best Fleetwood Mac song ever?
Oh, I gotta say that performance Silver Spring and the dance. So good. So good. I agree with that. Okay. So let's actually talk about what, what you've been invited on to chat about. What is a smart wall, and what are the technologies that go into making up this kind of virtual smart wall?
Yeah. So, in the, build up to the initial shutdown, Democrats and a lot of policymakers started referring to something called a smart wall, smart border security. They followed it up with saying things like drones and sensors. So what we take from, the term smart wall is essentially the use of technology and technical infrastructure to identify, border crossers and essentially using technology to complement CBP's,
border enforcement mission. Gotcha. So no physical physical wall per se, but it could be like a virtual, like, barrier paired with drones, paired with, like, cameras,
all that fun stuff. Yeah. You're painting exactly the kind of picture that, that they wanted. So essentially, things like sensors, aerial surveillance tools, like drones, certainly, automatic license plate readers. Oh. These are all different tools in the toolbox that,
could be deployed and actually are currently being deployed. So why are you worried about these? Don't these just all sound great? I'm kidding.
So, when we talk about technology deployment at the border, we're talking about a broad spectrum of different, technical capabilities. And what we tried to highlight in the build up to, the negotiation for the spending bill. And what we're thinking about now is how do we address or mitigate the harms from the particularly invasive technologies that we're concerned about? So these are, aerial surveillance tools. Drones certainly qualify. These are things like automatic license plate readers. These are the deployment of facial recognition technology. All of these, have a few things in common. They allow for the identification monitoring, of individuals. And that kind of warrantless monitoring, particularly, by the government chills activity, and CBP currently claims a lot of authority to use these tools however they see fit, wherever they see fit within the border zone, which is defined arbitrarily.
Yeah. What is the border zone? I actually read your post and I was surprised, you know, how large the border zone is. Right. So,
I have to go back and check the origin, but my understanding is, regulation was issued in the fifties by which the AG decided that the border zone or what would be a reasonable zone for CBP, or border patrol, to operate within would be a 100 miles. So the border zone is a 100 miles from the physical border. And that is an area, thanks to the great work of the ACLU, if you check out their website, 100 mile, border border zone, you can see it includes two thirds of the American population. It's a it's a really, really vast area.
And I would imagine when you talk about these technologies, you know, the border zone, there's a lot of people just going about their lives there. It's not just, you know, I guess in theory people trying to, you know, sneak into the country as some might say. It's probably just people who live there that this technology could then monitor.
Yeah. So we've seen, these kinds of invasive technologies, this kind of invasive surveillance deployed at the border for, you know, decades. Like, since 1997, we've tried to have, poorly acronymed ISIS, Yikes. Technology plan, for securing the border. And the reality is that when we deploy technologies at the border, it's people living near the border who are gonna bear the brunt of this surveillance. Government surveillance, prolonged government surveillance, it chills activity. People living at the border shouldn't have to think about, is the government watching me as I go to my doctor's appointment? Yeah. Are they watching me as I go to my place of worship? So these are the concerns that surveillance raise, and you're absolutely right. There are a lot of people who live within the 100 mile border zone Yeah. And also live particularly close to the border. And they face things like CBP's checkpoints, which stop them on their way to and from work, going to get groceries, etcetera. This is,
a surveillance infrastructure and dragnet apparatus that really impacts people's day to day lives. That's interesting. And then that data collected by DHS or Customs and Border, are there rules around how it's used then? I mean, if you're if all this data is being collected, whether it's your face, whether it's your license plate, are there any rules around its use? So
we have the privacy act, and there are rules on how there are rules in place, governing what happens when data is, is collected by DHS. However, when we look at them, they are they they grant the government a lot of latitude to retain data for a very long period of time and to share it very widely with Mhmm. State, local police, federal agencies. So there are rules, but they're really squishy.
Squishy insufficient rules. And I guess even, another kinda big question, does this technology even work? Like, does it address the issues that, you know, customs and border is trying to address?
Right. So, something that DHS has historically struggled with, and you can see this in their past, plans to deploy technology infrastructure at the border is deploying the right tools, the right place, and then assessing whether or not they are force multipliers or even effective. So, in 2005, a really famous SBI net was started,
and this was a plan Better name than ISIS.
Certainly a better name. But this was a plan to put up a physical fence along the entire Mexico US border. It would be complemented with technology like sensors, infrared cameras, things that would essentially detect when someone had hopped the fence. And after several years and $1,000,000,000 spent, they found that they weren't able to they weren't able to complete this project. They they, DHS had failed to oversee the contractor effectively. There are problems with the technology even working. Sensors, were, activated by weather, by perfectly lawful activity, like people living their lives, traffic.
So I assume those rabbits down on the border just hopping it all the time, you know? I
mean, the the sensors aren't gonna detect, like, is this the kind of specific activity that we're trying to deter? It's just gonna it's like, is it activated? Is there movement? And so they were finding that the the tools weren't helping them, and that CBP was actually being sent to locations where, you know, nothing was happening. It was just a weather related incident. So that project failed. There have been subsequent attempts to put together more specific tailored, technology security, programming along the border. Arizona's, known case study of this. But other tools, just as as a tool itself, has have not been proven very effective. So, famously, the IG reviewed CBP's use of drones. They've been using drones for several years. The IG determined What's the IG? The inspector general. Okay. Thank you. Determined that drones were dubious achievers. That's a literal quote from from from the assessment. Oh, no. Because they're expensive to operate. They take a lot of manpower. Yeah. Each drone took 20 people to operate. Something Wow. Some some figure a large figure like that, but, you know, then you're thinking, like, why are you even operating this thing? So it take it took it so they take a lot of manpower. They're very expensive. Weather precluded their deployment, and CBP couldn't assess how effective drones were in helping them, you know, apprehend people. I think, the report said something like, drones helped in two percent of apprehensions. It's a very invasive tool.
It seems like a high cost for only 2%.
Absolutely. Yeah. And that's that's a common problem essentially with deploying invasive technologies. It's like, okay. You have you're trying to secure the border. That's your stated mission. You're thinking these tools can help you. Well, not only are they incredibly invasive, but they're not even helping you, and they're expensive. A real assessment needs to be done, of of how every tool or how a piece of technology can actually assist CBP. And they need to assess, you know, metrics like how effective is this in helping us, with our mission, what are the attendant privacy costs and cost to civil liberties Yeah. And and assess, are there less invasive means for us to achieve our stated mission, which certainly seems like there are. Yeah.
So let's just go a little bit deeper on that. So the the funding is there, you know, obviously, they're gonna buy all this technology, you know, who knows what exactly it's going to be. Although you have some good ideas there of what it might be. What are the civil liberties questions you kind of started on them that should be asked? What are the things, whether privacy concerns or other concerns that, should be addressed before any of this technology is deployed?
Right. So we need, we need some specific guardrails to be put in place. First of all, the border zone is is too vast.
Two thirds of the population.
Wow. Yeah. It's, border missions and border security, is already very prone to mission creep, and just the amount of space within which, CBP is allowed to operate just allows these missions to creep out even further. So we need to scale back, the border zone. And I think it's possible that there's appetite in congress for that. So there's a geographic use limitation. Mhmm. There needs to be a limitation on CBP's sharing and retasking of these tools. So, with respect to drones, CBP is known to have shared drones with, state and local law enforcement agencies to retask it in support of missions unrelated to border patrol. If congress is authorizing CBP to use a tool for its specific mission, shouldn't be resharing it, resharing tools, with with local and state officials without serious limitations on what they can do with the technology. We need, restrictions on better restrictions on data retention and sharing. Currently, the rules are far too lax, and, I those those are definitely ripe for some reform. And so so what these guard rails would do is they would protect against, warrantless, persistent surveillance, which is what some of these tools that we've been talking about facilitate.
That's the concern. Yeah. That would be really crappy to live
in a world like that. Yeah. I mean, who wants to just live your, like, live your life and then there are drones up above you. Your car your location in your car is being tracked every time you pass an automatic license plate reader. You have to engage with CBP frequently.
That's not a world I think anyone wants to live in. Yeah. And we can't even tell people to move away from the border if it covers two thirds of those countries or the population. I know technicality. Well, Mana, that was very, very insightful. Thank you so much for joining. Mana has a great blog post up on CDT's website if you wanna read more about this. Thanks for joining, Mana, and enjoy Fleetwood Mac and the legendary miss Stevie Nicks. Thank you. Thank you. Lovely to be on. I'm definitely gonna enjoy Fleetwood Mac tonight. When the controversial allowing states and victims to fight online trafficking act, that's a long one, or FOSTA became law, CDT warned that it would lead to the censorship of legal speech. Despite its important goal of trying to curb online trafficking, the law, like many before it, took a sledgehammer to a core Internet tenant that has allowed online speech to flourish without demonstrating that it would in fact address the actual problem. This law, of course, is being challenged in the courts, and our free expression director, Emma Alonso, is here to share the latest. Welcome, Emma. Hey, Brian. It's great to be back. Yeah. It's It's both of our, like, return to the podcast after sabbaticals. A great CDT policy. Work here five years, you get six weeks off. Yep. And apparently, go to New Zealand with it. So Brian and I both decided that was the vacation destination for, January. Yes. And that's the next podcast episode on New Zealand. No. It's not. Sorry if that's what you're looking for. All tech policy here. So, Emma, remind us what is FOSTA, or I suppose also it was called SESTA at some point, and then it became FOSTA. And tell us what it was and why CDT opposed.
Sure. So you may remember FOSTA and its companion, FOSTA, from, this time last year when congress, after kind of many years of of taking a look at the issue, and some, intense work starting about September 2017, passed a law that was aimed at restricting the spread of information about human trafficking, particularly advertisements, on online platforms. The goal at least was sort of framed as cracking down on websites that host, content that that facilitates or enables trafficking. But what the law really did, kind of in in actuality was create really broad new federal criminal and civil liability for hosting content that promotes or facilitates prostitution, which is a really pretty vague standard that could encompass a lot of kinds of speech. Mhmm. It also created some new exceptions to section two thirty, the statute that protects, intermediaries, so content hosts, search engines, access providers from liability for content that their users generate. FOSTA was the first law to, actually amend or change section two thirty since it was passed back in 1996.
Wow. So before we get to that, because I'm definitely gonna want to tell you or hear a bit more about the challenges that have taken place and why you maybe think that this is the first one to have actually passed. The outcome that we were predicted, online censorship, has it happened?
Yes. Pretty much immediately after the passage of the law, we started seeing different websites, you know, shutting down or, major websites closing down particular threads. Two of the highest, highest profile sites to do this were Craigslist, which shut down, its entire adult services section which included the missed connections section, and a lot of other kinds of of ads that it would host, and also Reddit which closed a number of subreddits that, you know, it was concerned, might be viewed as falling under this really broad and amorphous definition of facilitating or promoting prostitution, by kind of providing information either about sex work, about prostitution, advocating for legalization, or what have you. So pretty much immediately we saw both high profile sites and then other sites, you know, devoted to harm reduction for sex workers, people trying to provide health and safety information, people, you know, providing places where, you know, people who are engaged in sex work can share information about clients that they've had as a way to try to encourage, you know, safer experiences and and share information about people, to to stay away from. You know, there's a whole host of content and, and kinds of sites and services that saw the law passed and before it was even actually kind of in effect, took kind of immediate action to, to shut down rather than risk this really substantial new,
threat of liability. Yeah. And, I'm as you're talking about that, I'm thinking back to the event that, we worked on together, the Future of Speech Online back in December. And we had Maggie Mayhem there who advocates for sex workers rights. And she was talking about how a lot of these these forums, just like you were describing to share experiences, whether health, you know, issues or, whether health, you know, issues or, particular clients that were abusive to avoid, those were getting shut down and really the impact it had on people's lives. So, folks should take a list either watch on, watch the videos online. We have it from the Future of Speech online. Her name is Maggie Mayhem or we had her in the podcast. So, Emma, what is happening then? This is now being challenged. We filed a brief and it's complicated. Explain it to me.
So, back in the summer, a a group of plaintiffs, including the Woodhull Freedom Foundation, Human Rights Watch, archive.org, and two individuals, Eric Kosick and Alex Andrews, together brought a constitutional challenge against FOSTA saying that this law, violates the first amendment. It is vague. It's overbroad. It's not narrow narrowly tailored. Pretty much every way you can violate the first amendment, you know, FOSTA arguably falls prey to. This case was brought, at a at the district court, which dismissed the case for lack of standing. They said that the, the plaintiffs couldn't show, a credible threat, that they might be prosecuted under this law and that they didn't really, therefore, face any kind of chilling effect on their speech, because there was no real risk of them getting prosecuted. This so this that dismissal of the case is now being appealed, and that's the Gotcha. That's the case that that we just filed a brief in. And I'm so that's it's before the, Court of Appeals for the DC Circuit, and I'm really hoping that the DC Circuit, steps in and and overturns that dismissal of the case, because the district court got things wrong in a couple of ways. One, the the doctrine under the first amendment for standing, for determining when plaintiffs have a the ability to go to a court and say this law chills my speech. It needs to be stopped. It needs to be reviewed, maybe struck down. That standard is that courts are supposed to, kind of unless there is really credible evidence that there's no credible threat of prosecution, Courts are kind of supposed to on the side of what the plaintiffs are saying, that they fear prosecution, that there is, and therefore, there is a chilling effect. The district court, really kind of decided things giving heavy weight to the the government's perspective in this. So the, Department of Justice, was defending the lawsuit and they said, you know, we're not planning to prosecute any of these plaintiffs so they have no credible threat. But what this also fails to recognize is that Foster did more than just further empower DOJ. There are federal criminal prosecutions that can be newly brought under Foster, but it also enabled state attorneys general and civil plaintiffs to bring certain kinds of claims against people. And Department of Justice can't kind of promise on behalf of all state attorneys general and all civil plaintiffs that none of these plaintiffs or or other people will threat this or face this kind of threat of prosecution. So for the district court to have kind of bought in so much to the the government's presentation of the case and, to have sort of literally dismissed the the claims that the plaintiffs were bringing, that they genuinely already had faced, chilling effect on their speech, we think was wrong and hope that the court of appeals overturns that. Okay.
So I'm gonna ask you kind of at the end how you how you think this may play out. But, maybe it would be useful for you to inform that by a bit of history, which I know is a fair amount of what you, wrote, you and your colleague, Liz, wrote in our brief. Tell us a little bit about the history of section two thirty and maybe some relevant bits that may come into play
Sure. When when deciding this. Yeah. So, our brief was one of four, amicus briefs that were filed on behalf of the plaintiffs. There was also a great brief from, engine advocacy, Reddit, and the Copia Institute about sort of the effects this law will have on website operators from, you know, from their perspectives running websites. There was also a really great from the the Freedom Network, and a number of other organizations that advocate for sex workers and survivors of trafficking, and another a number of other interests that really kind of spelled out the, the harm that this law is immediately having on all of their constituents, and, and a number of people. Our brief really focused on the kind of the history and, background on congressional and state attempts to restrict certain kinds of information online. And no surprise, especially to people who've been kind of following the history of of Internet regulation in The US, other than copyright, you know, content having to do with sex has probably been the, you know, the primary focus of what, congress has tried to regulate over the years. Mhmm. But, you know, long story short, it's not had much success at all. It is very difficult to craft a restriction on speech that comports with the first amendment. You know, the the first amendment And that's whether online or off. Right? Online or offline. Right? And the the first amendment's prohibition that Congress shall make no law bridging the freedom of speech is a a really solid baseline to start from of presumption against, you know, whatever whatever this restriction is. There are obviously different cases where, limitations on speech can go forward, but most of the times that we've seen, congress pass things like the Communications Decency Act back in 1996 which tried to make website operators liable for enabling, minors people 18 to access indecent content think pornography stuff that's lawful for adults to access but states can limit minors access to. The the CDA tried to make website operators responsible for preventing minors from accessing this lawful for adults court struck that debt. The Supreme Court struck that down as a violation of the first amendment in the landmark case Reno versus ACLU, the first big Internet free speech case in The US. And really, it's a great case that has set the kind of the the benchmark for our legal systems understanding that the role of intermediaries, the role of website operators, content hosts, domain name providers, is really intrinsic to how we as people can use the Internet for our enjoyment of our free speech rights, and that efforts to kind of leverage those intermediaries' control over speech can have a real, chilling effect and can essentially stop people from being able to to speak and access information online. There were a number of other cases both the federal level or laws at the federal level and in the states, but you know the long and short of it is most of the regulations that we've seen trying to restrict online speech, tend to be vague because they're it's hard to define, you know, draw that line between unlawful speech in a way that's clear enough to speakers let alone intermediaries about this is what is prohibited and the conditions under which it's prohibited, and we have an overbreadth problem very often that the the ways that laws try to go about,
stopping the spread of illegal speech also sweep up a lot of protected speech as well. And, you know, kind of back to the example that you gave earlier, a lot of the platforms that shut down or, you know, just talking you could be an advocate for sex workers' rights and get have your platform or your forum shut down because of this. But it it you do raise kind of the interesting question, you know, what I our president has said before. You know, liability does not mean some responsibility. A lot of people would probably agree with that. How do platforms make decisions about what content? If it's if not this channel Mhmm. Then, you know, how do they make some of these decisions about? Either content that they don't want up there or don't feel, you know, as part of the community that they have? Or Yeah. No. That's a great question. It's one that's being really actively debated in
public and, you know, in popular press, more so now than I think any time before in the history of the Internet. It's it's amazing to me how quickly content moderation has moved from a niche, extremely wonky issue to something that everybody talks about in the context of, you know, every policy debate. And but I think what the, you know, the answer is it's not easy. Right? Running a platform really of of any kind of scale or size requires making a lot of different sorts of trade offs, whether it's, you know, wanting to have a really prescriptive content policy that's trying to be as clear as possible about all of the stuff that is or isn't allowed on the platform or having a very general policy that has some, you know, high level principles that kind of guide what you do and and doesn't get down to the the nitty gritty of, you know, defining exactly what counts as, hate speech or harmful content or something. You know, different platforms take different approaches with whether they use kind of paid moderators or volunteer moderators, both of which raise really interesting questions about what is the job that moderator is doing, what are they exposed to, you know, is that can they do this job in a way that maintains their their health and safety? These are really interesting challenging questions but one thing I think we can see including from some recent examples is that when, a content host, a website, a platform is feels like they need to try to enforce a really kind of complicated or nuanced policy a lot of times over breadth is one of the the the responses. A lot of times What's an example? The so so a good example of this right now is, a very recent one probably saw the headlines about, Pinterest taking, a really kind of aggressive stance against, anti vax content content that, you know, falsely, claims health risks of of vaccines or or discourages people from getting vaccines using, false information, you know, Pinterest has decided that at least for now their their approach to dealing with the fact that there was a lot of anti vaxx content on their site is to not allow people to search for content at all related to vaccines, whether it's factually accurate information or anti vaxx content. Now if you go to to Pinterest and you search vaccine, there are just no results. Wow. This yeah. It's a pretty broad response, but I think reflects a couple of things. One, that it can be really difficult for a platform to wade into these kinds of content moderation issues. Right? There's no law requiring Pinterest to do this. In fact, the law, section two thirty, kind of shields them for for their decision to do this, for their decision to say, we don't want any of this kind of content to show up in our search results. For their they also have a policy that enables Pinterest to take down content that might lead people to, I think, like, harmful health effects, something like that. The law shields them for for making those decisions and kind of enables them to, in this case, try to get a handle on the misinformation problem on their platform by taking this really broad best approach. But what it does is also mean that nobody's gonna be able to go to Pinterest and find, good information about vaccines, helpful information, stuff that clarifies misconceptions or anything. Now for a platform like Pinterest, that may not be what most people think it's primarily there for. Right? It's, you know, it's not the Mayo Clinic. It's not, necessarily anybody's top source for really scientifically accurate medical information. But imagine, you know, a search engine Yeah. Or, or even just a a really big general interest platform like Reddit. A site like that applying this platform would really start kind of interrupting the ability for their users to access information they really do want to access there. Yeah. So I think it's a good example of how certain kinds of broad brush approaches to to dealing with, you know, challenging content or or content where it's difficult to to draw lines and maybe politically sensitive for a platform to draw lines. One response is to take this broad brush approach,
but it that kind of approach has a lot of its own downsides too. Yeah. And it might work for some platforms and not for others as you said. I mean, ones where it could be far more harmful than helpful potentially. So last question. Let's just go back to our, the brief that you filed and and FOSTA. How do you think that plays out in the courts? Do you think the things that they're gonna decide it the way you'd like, and what happens after that then? Well, hope springs eternal.
So so what the plaintiffs have asked for, what we're supporting is, for the the court of appeals essentially to allow the case to go forward to say that, the plaintiffs do in fact have standing to challenge this law. Plaintiffs are also asking for an, an injunction on the law, so to kind of to to pause it to make sure that neither federal nor nor state prosecutors or civil plaintiffs can bring cases under this law, until there's a chance for a full kind of hearing on the merits where all of the constitutional claims are thoroughly vetted. You know, and that kind of process can take a while. So having a a preliminary injunction against the law keeps it from having, this potential chilling effect while that case is still going forward. So hopefully, that's what we hear from, from the Court of Appeals. I wouldn't be be surprised to see this one go all the way to the Supreme Court. Alright. So that's a big stay tuned. Yeah. We'll keep you in in business.
Well, Emma, thank you. Always a pleasure. I always learn something when we chat, and it's great to have this be the first one I recorded post sabbatical. Yeah. Thanks so much for chatting with me. And that's it for this episode of Tech Talk. If you wanna read more about either topic we discussed today, you'll find some lovely prose from both Mana and Emma at c d t dot org. I'm Brian Waslowski. Thanks for listening.
