Human Voices, Episode 3: Big Tech’s Control Over The Media, ft. Dipayan Ghosh

Hosted By Bastian Purrer

October 28, 2020

Welcome back to Human Voices, HumanID’s podcast talking cybersecurity, privacy, and news in tech! This week’s focus is on the intersection of digital media and civil rights. Dipayan Ghosh is a senior fellow at the Harvard Kennedy School and lectures at the law school. He was a technology and economics policy advisor in the Obama White House and a global and private policy strategist at Facebook, both sort of around the 2016 elections. Recently, he released a book “Terms of Disservice: How Silicon Valley is Destructive by Design”, which is a critique of Big Tech and how they sold out the consumer society and, partially, American democracy. Ghosh urges for a radical reform, focused on competition, privacy, and transparency.

You can listen along to Human Voices right here or on Spotify, Apple Podcasts, Anchor, or ListenNotes! The below teaser has been edited for clarity and brevity by Ariana Garcia. 


humanID: What brought you to write this book specifically? What’s the bigger goal?

Dipayan Ghosh: I think what I was most concerned about, and what I think humanID is concerned about, is that when we think about the digital economy today, it’s really dominated by a few companies. In particular, I’m thinking of the typical American consumer experience of the internet. It’s all about Facebook, Amazon, and Google. I think that part of the reason for that is that those three companies have monopolized the consumer internet industry. I think we need a reform agenda. The time is right.

humanID: What happens when we leave Facebook in control of policing what is right and what is wrong? Isn’t there a danger to give Facebook the sort of rulebook and be the lawyer and be the judge of what is right and what is wrong to say? And doesn’t that that doesn’t that endanger freedom of speech?

Ghosh: There’s no doubt it does. I’m not necessarily suggesting that Facebook should be the police officer, a cop on the beat of speech regulation. But the absence of an appropriate cop on the beat now, in the lead up to the 2020 election. I think we do need Facebook to step up just as other companies have done.

humanID: And do things like what Reddit did and ban The Donald?

Ghosh: Or even what Twitter has done in flagging some of Donald Trump’s tweets. Facebook, of course, didn’t take any action. It’s quite problematic when you have one company that’s essentially putting democratic interests over profit interests, but the rest of the industry is not quite following suit, especially the most important company in the media space right now.

humanID: But these are things that are not illegal to say in America. So why should they be illegal on a platform?

Ghosh: We have a line in the United States constitution, section 230 of the Communications Decency Act, which essentially gives technology companies immunity from regulatory intervention on two fronts. First of all, they have the freedom to leave up any user generated content that they wish and not be liable for what that content says, even if it’s totally egregious. Second, if they actually do wish to take down content, because they may view it as egregious, they can do so without any fear of intervention from the government. So they have this free pass on both sides.I nstead of giving tech companies kind of a free pass on both fronts, what I’d suggest is that they should have regulatory scrutiny on both fronts.

humanID: So I tend to agree with this. There’s a danger in not giving any liability for moderation and deletion and the algorithms. But, something like shadowbanning is just an incredible, dangerous form of censorship.

Ghosh: And of course this conversation is, as technology conversations do, go straight back into this question of What is what is good content? And what is bad content? To which my answer in my book is: Don’t even look at content. We can design democratic ways to determine what should be socially acceptable in the political realm or the realm of hate or any offending content, extremist content, exploitative content, violence, hate and disinformation. All of these, in my view, are forms of offending content. But where does that red line stand? I don’t know. I’d suggest that we should crowdsource that answer from society because no one individual’s answer is going to be right, because this is such a partisan issue.

humanID: You’re right. It’s amusing that we end up with this topic, again, because at HumanID we’re working under the hypothesis that if everyone has just one digital identity then we can be held accountable by other individuals as well as platforms and we don’t need to censor and can apply the same standards we do offline. We also think that kind of looking at the content is actually not necessarily the right way, rather, looking at the identity.

Does anything give you confidence that a platform like TikTok, that is ultimately under the control or indirect control of a foreign government, doesn’t abuse access to our teenagers?

humanID: TikTok clearly has to follow the guidance of a foreign government. How can we protect American or European democracy against that being abused, to have voters turned against us?

Ghosh: I think TikTok is a very interesting company. In the past few days, a particular Reddit post has been circulating everywhere. It says, essentially, that TikTok, based on this expert’s, Bangerlol, reverse engineering of the mobile application, he’s found out that TikTok collects orders of magnitude more data than Instagram, Whatsapp, Facebook, and potentially other applications. This is a very high bar given how much data American applications collect. He also makes a series of accusations and the most egregious of which is that TikTok has been doing certain things to try to conceal all of this, and even design the app to work against any reverse engineering.

To be honest, I wouldn’t have expected anything else based on a reverse engineering of the application. You know, TikTok is not governed by a democratic regulatory body. It’s a Chinese application and probably has connections to Chinese government and works according to a different set of conditions, including commercial and cultural ones.

humanID: Should America block TikTok?

Ghosh: I think so yes. To put it another way, I think America should have a very strong privacy law. And I’d argue that TikTok completely violates this theoretical privacy law, and therefore should receive an injunction and be blocked from the United States until it corrects itself.

A solution like humanID has tremendous potential because it essentially rebalances the distribution of power from the monopoly platform into the hands of the user.