Inside the Covered Wagon: Privacy, Anonymity, and the Future of Open Societies
Part I of a multi-part series on the tensions between emerging technology, privacy, and free and open societies.
Emerging technology has a way of encapsulating multiple policy issues into a single debate. For example, the “TikTok” ban has led China Hawks, privacy advocates, trust-busters, and censors in a murky quagmire of a policy debate that leaves it unclear what “the real reason” the platform should or should not be banned. While each camp has merits to its “ban or not to ban” arguments, the multi-faceted nature of the problem often leads to issues being conflated with one another. It is sometimes helpful to compartmentalize the debate and think through the issues and merits of proposed solutions on their own.
A few weeks ago, I covered the TikTok debates as an example of the asymmetric advantage authoritarian states have over open societies. In that piece, I began to argue that one of the asymmetric advantages is that the open society does not always know the identity of the speaker, thus making it difficult for the electorate to evaluate the speaker’s argument.
Can we solve the asymmetry of information our society faces with foreign influence in ways that do not restrict the information itself? In the asymmetry of anonymity, the solution may be simple—identifying the speaker. The problem is not what China tells Americans through TikTok and other platforms but the anonymity with which they speak and the lack of anonymity through which Americans listen.
When the Athenians stood before the Melians, the Melians knew precisely who they were and could effectively weigh the consequences of the arguments and propositions the Athenians laid out. If the Melian dialogue had occurred under today’s circumstances, not only would the Athenians have owned the forum where they spoke, but they would have done so in a way that would have obscured their identity to the point where it was unclear who was issuing the ultimatum. The opinions of foreign nations are important pieces of information for citizens when making a policy choice, but that information must be complete.
To argue for a solution that involves exposing the identity of speakers on the internet is controversial, to say the least. The trade-offs and consequences for such a policy would affect different groups disparately—whistleblowers, journalists, and marginalized communities could all be in danger should their identity be compromised on the internet. Yet the policy solution does not have to be draconian or binary; we do not have to pick between anonymity, privacy, or full-frontal online exposure. Ignoring the fact that bots and foreign agents have unfettered access to our (non-anonymous) electorate is not a problem that can be ignored either. Reducing online anonymity may or may not be the right solution, but it is a solution being discussed, and given the possible ramifications, it should be thoroughly vetted.
Below is a collection of stories and resources on privacy and anonymity I have collected while writing Pioneering Oversight. It contains the good and the bad across as many sectors as I could find. It is meant to be a jumping-off point, a bit of a primer, to start a multi-part series on emerging technologies’ impact on privacy and anonymity and the effect it has on open democratic societies.
The Basics
An overview of Privacy vs. Anonymity from Proton, a well-respected technology company offering privacy solutions.
Anonymity is not the same thing as privacy, and confusing these related but different concepts can weaken your online threat model.
Anonymity – Keeping your identity private, but not your actions. For example, using a pseudonym to post messages to a social media platform.
Privacy – Keeping some things to yourself, which can include your actions. For example, messaging friends privately so they know who sent the message, but only they can read it.
Both concepts are different from security, which can often add to the confusion.
Security – The precautions used to keep you safe. What exactly “safe” means can very much depend on your threat model.
Articles that challenge the notion that anyone is really anonymous online and how to become anonymous if you really want to be:
The Markup’s feed of privacy investigations
Reversing Privacy Risks: Strict Limitations on the Use of Communications Metadata and Telemetry Information:
[This paper lays] out the extent to which “non-content”—the hidden parts of Internet communications (aspects the user does not explicitly enter) and telemetry—are highly revelatory of personal behavior. We show that, privacy policies notwithstanding, users rarely know that the metadata and telemetry information is being collected and almost never know the uses to which it is being put.
Second, we show that consumers, even if they knew the uses to which this type of personal information were being put, lack effective means to control the use of this type of data…This is especially the case for telemetry, where the ability to understand both present and future use of the data provided from the sensors requires a deeper understanding of what information these devices can provide than anyone but a trained engineer would know.
Information Warfare
Dark Corners of the Web Offer a Glimpse at A.I.’s Nefarious Future
In the hands of anonymous internet users, A.I. tools can create waves of harassing and racist material. It’s already happening on the anonymous message board 4chan.
Disinformation Is a Threat to Our Democracy, By Barack Obama
China
How 2023 marked the death of anonymity online in China
In April last year, Chinese social media companies started requiring all users to show their location, tagged via their IP address. Then, this past October, platforms started asking accounts with over 500,000 followers to disclose their real names on their profiles. Many people, including me, worry that the real-name rule will reach everyone soon. Meanwhile, popular platforms like the Q&A forum Zhihu disabled features that let anyone post anonymous replies.
Each one of these changes seemed incremental when first announced, but now, together, they amount to a vibe shift. It was one thing to be aware of the surveillance from the government, but it’s another thing to realize that every stranger on the internet knows about you too.
AI-tocracy
Recent scholarship has suggested that artificial intelligence (AI) technology and autocratic regimes may be mutually reinforcing…We first show that autocrats benefit from AI: local unrest leads to greater government procurement of facial-recognition AI as a new technology of political control, and increased AI procurement indeed suppresses subsequent unrest. We show that AI innovation benefits from autocrats’ suppression of unrest: the contracted AI firms innovate more both for the government and commercial markets and are more likely to export their products; noncontracted AI firms do not experience detectable negative spillovers. Taken together, these results suggest the possibility of sustained AI innovation under the Chinese regime: AI innovation entrenches the regime, and the regime’s investment in AI for political control stimulates further frontier innovation.
Pornography
How a ‘Digital Peeping Tom’ Unmasked Porn Actors
“It’s like the secret identity of Batman or Superman. You’re not supposed to know who this person is, they didn’t want you to know, and somehow you found out.”
Meet the 15-year-old deepfake victim pushing Congress into action
You’ll Soon Need to Show ID to Watch Porn Online
^^ This article is almost a year old. Florida recently became the 8th state to enact a version of these laws.
Surveillance
How scientists traced a mysterious covid case back to six toilets
In an effort to contain Covid, ethical questions about privacy create tensions with public health:
But how far should they go to find these people? That’s still an open question. O’Connor can imagine a dizzying array of problems that might arise if they did identify an individual shedding one of these rare variants. The most plausible hypothesis is that the lineages arise in individuals who have immune disorders that make it difficult for them to eliminate the infection. That raises a whole host of other thorny questions: what if that person had a compromised immune system due to HIV in addition to the strange covid variant? What if that person didn’t know they were HIV positive, or didn’t want to divulge their HIV status? What if the researchers told them about the infection, but the person couldn’t access treatment? “If you imagine what the worst-case scenarios are, they’re pretty bad,” O’Connor says.
On the other hand, O’Connor says, they think there are a lot of these people around the country and the world. “Isn't there also an ethical obligation to try to learn what we can so that we can try to help people who are harboring these viruses?” he asks.
‘Cognitive Liberty’ Is the Human Right We Need to Talk About—how new brain implants create science-fiction type privacy concerns.
How Citizen Surveillance Ate San Francisco
March 25th Pioneering Oversight discussing privacy in vehicles.
Policing
Cops Used DNA to Predict a Suspect’s Face—and Tried to Run Facial Recognition on It
Clue in the windscreen may be key in legal challenge to Queensland’s AI traffic fine system
The state’s traffic camera revenue is soaring, but the technology isn’t faultless. Amid several legal challenges, it’s not just drivers raising concerns
AI to help churches and castles fight wave of graffiti and vandalism
War
Ukraine’s use of facial recognition system.
