Ep. 168: Consumed in moderation

I, V, M. and we are all here for social validation, pretty much like a topic for today, Facebook. There are 4, there are 4 amachos going to do this so you can actually say I'm charm, amachar. That is only professionallyomaite here.

Okay so this is Tony and I'm joined by, self is stuck. And we keep staring at rain outside window. Oh wow, okay, we are not staring at Ma Ran, who is not picky? I am calling an embargo on this joke.

As funny as it is, because it's at least the fifth time that we have done this. I think it's time to make jokes on Sree Kid's name now. Sree Kid? Yeah.

Okay. So before we... Sree Kid wins lit. Oh wow.

So Titanic mistake. Okay. So before we get into Facebook and all of these virtual realities, I have a question for all simplified hosts. And this has been triggered actually by...

After the last recording, I met my neighbor last week and it's like, hey, are you the same Tony who's un-simified? Are you serious? Yeah. And I'm like, wow.

No, this is actually at my home. So I was like, pleasantly surprised. And she said, I was not sure. She was judging me.

So yeah, any brokers listening to the podcast, I was looking for other apartment to move into. And hopefully the brokers would have also simplified. Yeah. But shout out to Chico.

Thank you for being a supporter. Oh, nice to meet you. Nice to meet you. Nice to meet you.

No, I did. But my question for you guys is, have you had real life simplified experiences? Are you coming up to you? Yeah.

So this is very strange. Okay. So I was just here, weekend of 2019 and a bunch of us were up front waiting for Opeth to begin the headlining act on day two. And I was just talking with somebody who was over there.

And somebody behind me actually came and I said, are you by any chance Chico runs simplified? I said, yes. How did you know? Is that I recognized your voice?

I don't know. Yeah. Amazing. Yeah.

So something similar happened to me as well. I was, we had a presentation of one of the studies that we had done over here and I was presenting to basically a room full of clients. And after the presentation was done, this guy comes up and he's like a pretty senior brand manager client over here. And he's like, are you, are you friend of course, as it began, are you a friend of Chuck?

I'm like, yeah, you're on simplified. Aren't you? I'm like, yes. And he's like, no, I didn't know that before.

But when you started talking, I realized I was like, okay. I think, are you a friend of Chuck with respect to you is basically another way of saying I really don't know how to spell your name. Yeah, that's true. I mean, well, that's fair enough.

Or pronounce. That's why when the Chinese people over here say, I'm sorry, your name's really difficult to spell. I was like, huh, I'm not. And he's mainly her bag.

And Narin is not here, but I can answer on his behalf because he has said this on the episode several times before where he has gone to, when he was in the US or something, when he went there to, I think it's the sun or work or something. He, I think a couple of people in the bar or something came up to him and asked, are you, I mean, they, no, no, it happened to his son actually. He actually, so he's not. So they figured out that he was a Shenoy and said, by any chance, are you related to Narin the Shenoy or simplified?

Okay. I think the report is that. Wow. Yeah.

So Narin's report was that his son was like, apologetically, he tried to get out of there. That's not a bad idea. Right. So, but now we suspect that beneath that veneer of, of shyness, Narin is proud of what he has done.

Yeah. Very nice. That is okay. Okay.

That's the episode. Thank you. We have decided to convert simplified into a therapy session for the fourth course. Thank you.

Yeah. Next week. Simple for yourself. I know.

Social validation over. Okay. Now we've got over the validation, we'll get into the... Social, we'll get into the social.

The message. So, exactly. We'll be talking about Facebook and specifically its content or discontent as things maybe. Wow.

And honestly, I didn't know where to start. So, obviously I went to the Wikipedia page called Criticism of Facebook. Okay. There is a page.

There is a section like... Yeah. And we wait. This is what the...

There is a notice on that section where admins put it up. No. And this was put up in June 2017. It's still there.

This is what it says. Criticism of Facebook. This article may be too long to read and navigate comfortably. The readable process is 106 kilobytes.

Please consider splitting content into sub-articles, condensing it or adding sub-headings. And this page alone, which is Criticism of Facebook, has 20 sections and 94 sub-sections under it. So, I promptly close the page. I have Criticism of Criticism of Facebook.

So yeah, I think we should just go around the table and start by talking about what are your views on Facebook and its content moderation with docs. And then we can get back to... Wow. Okay.

This is a loaded... Yeah. I don't know. To be very frank, I can't think of an answer.

I mean, it's very easy to blame them because they're so visible out there and Zuck is doing nothing. But on the other hand, you can also... I mean, what can they do? Should they shut down entirely?

There is one school I've thought about it before, whether it's a good thing. I think Scott Galloway, the very outspoken New York University professor, put it very nicely when he said... I think Facebook has been a net negative for society. If you take Facebook away, yes, they will miss some parts of things.

But we have now found other ways to connect with each other. And for many of us, we don't even use Facebook that actively anymore to be in touch with who we care about. So if you take Facebook, many of us might not even miss it. And all the evils that it has, whatever it is, spreading misinformation or whatever it is, will be taken away.

So I don't know. Part of me that wants to sort of side with that, it's a radical band-aid type of solution. But I don't know the misinformation to find its way onto other platforms. One thing that I do feel is that they're not doing enough.

That recent Supreme Court thing that they did is one step in the right direction. I think it's genuinely a very, very innovative thing that they want to do. And I'm guessing we'll talk about that at some point over here. But taking stances like we are going to continue political advertising, even if they're full of lies, I'm not a big fan of that.

It's really bad PR for a company that doesn't need any more bad PR at the moment. Okay. Yeah. So again, I feel again, yes, there are arguments and chuck what arguments chuck made up pretty valid.

What I feel at the end of the day is that like it or hate it, Facebook is a business. It's a business that's working to kind of generate coin for itself and people will be liking its product. Now, I think people are voting with their feet like Facebook, the platform itself, not the company. But Facebook itself is using users regularly because I think the model is getting outdated and people are getting very afraid.

It's positive or negative. They're just getting very afraid. And I think on their level, Facebook is, yes, they're making efforts and they're doing the lip service. And I think this is the most hilarious, but I think somewhere I'd write about how in modern history, the guy who has publicly apologized the most is probably Mark Zuckerberg.

And so, yeah, so I mean, he knows that he can apologize and get away with shit. And at the end of the day, yeah, they're obviously not doing enough and as a responsible, say, for example, if they were like a responsible government body, then you'd expect them to do more. But they are Facebook. I mean, they're doing as much as they are expected to as a company.

And I mean, he's doing as much as he can get away with that. I think they're doing as much as they can. They're doing as much as they can get away with. I think that's important.

I think that's an important distinction. I think at some point in the future, see, advertisers are also becoming smart. At some point, there will be a tipping point where their earnings are going to reduce and that is going to be a pivotal moment. I think like, till now, the juggernaut is just going forward saying, yeah, I miss information or whatever, but people are still pumping money into the platform.

My people are advertisers. Advertis are getting smarter. They're pulling advertising out of YouTube, for instance. They don't want to be associated with misinformation.

I think eventually there is, I don't know whether it's going to be this year, next year, next 10 years, but I feel it's going to happen at some point. There are going to be better ways to reach out to people. I think all three of us have our own views about digital marketing itself, let alone digital advertising on a misinformation based platform. Yeah, if that tipping point happens, everybody's going to sit up and take notice.

Okay. Yeah, I think I hold ambivalent thoughts on this. I'm right now. It was in the deep end.

I'll come out of it, but I'm wearing a t-shirt that says, Descent is a safety value democracy and also wearing a sock which has Silicon Valley on it. So like Descent in the streets, Silicon Valley in the sheets, kind of. Wow. I think over a period of time, I've had many arguments with people where I've held a contrarian view and I am inclined to be more kind to Facebook.

And I think just like with the Mariday episode, I have a contrarian view to the popular narrative. I think I'm one, Ravi Shastri is amazing podcast away from turning into an insufferable uncle like Manu al-Sif. Wow, let's all just take a minute to appreciate the fact that in a panel universe, there could be a Ravi Shastri podcast. It was just amazing podcast.

Please reduce volume by about 50%. Yeah. But so the reason I am inclined to be more kind towards them is so Eric Schmidt of Google had famously said this, that the internet is the first thing that humanity has built, that humanity doesn't understand. It's the largest experiment in anarchy we've ever had, which can also be a tagline for simplifying us.

But essentially, if you look at any type of platform, whether that's YouTube or Twitter or any of the other biggies, I think Facebook does a better job than the others. At least they have a process which is more streamlined and you have some visibility on it. And all of them are black boxes when it comes to content moderation. But Facebook in some sense is a little better organized.

And we'll come to the Supreme Court part. But I think there's a really nice radio lab episode called Post No Evil, which essentially talks about Facebook's history with content moderation. We'll link to it from the show notes, but it's quite nice. I like what you said about the modern, and yeah, I'll try my best to be kind to Facebook.

Oh yeah, they are very easy punching bag, right? I mean, just small things. If anyone's listening to this episode and ever wondered why they're getting the ads that they are, Facebook actually makes it more transparent than many of the other platforms. I would say the most transparent.

All you need to do is click that. Why am I seeing this ad thing? And you will find the reasons why this brand wants to target people who are the stage, stage, stage, whatever it is and has interest in whatever, which is why you're seeing this. I think that's reasonably transparent of them to that.

I think they need to do that. And they don't listen to your microphones. Oh, yeah, they don't. But I think the founding principle of Facebook, it's one of the things I hated for is that they are greedy capitalists, but so is everyone.

That's the reality of the world today. But the founding principles of Facebook is based on free speech and democratization of information. And I think the reason that we have, let's say, active community of people exchanging ideas, a lot of that has to do with Facebook. Fair enough.

Even if not, Facebook per se, we all have WhatsApp groups. And this episode is going to be shared with all of us on Facebook. So it's not like... Absolutely.

And I think basically in 2018, which is when a lot of, it's basically over a decade since Facebook has been founded, a lot of things have happened. Cambridge Analytica, all of that. And Mark Zuckerberg wrote out a blueprint for content governance and enforcement, which ultimately turned into what people are referring to as the Supreme Court for content today. And that is what we'll be discussing.

But one of the things that he lays out in the document is about polarization extremism. And I'll just quote, he says, one of the most painful lessons I've learned is that when you connect two billion people, you will see all the beauty and ugliness of humanity. Which I mean, is very true, right? Like, there are 2.4 billion people on Facebook, man.

And people can say whatever they want to. How do you control that content? Yeah. Especially because there is no agreement on the right approach, right?

Yeah. Yeah. Like, and I mean, by being liberals in early 2020, we know that no two liberals can agree with each other, right? I disagree with that point.

No, think of the whole Kunal Kamra incident. Kunal Kamra incident as a piece of content, right? You won't find two liberals or like extreme free speech supporters having the same point of view on that. Now, if you're moderating a platform of content, which has so many of these, how do you come to a consensus on what content is allowed, what is disallowed?

And of course, these 2.4 billion people are not based in one country, but they are from different cultures and stuff like that, right? So what Facebook has always had is something called community standards. Have you ever had a post of yours removed from Facebook? No, actually.

Yeah. I mean, talk about, I talk about Progmetaller. Yeah. So I was actually once banned from Facebook for three days.

Wow, really? Yeah. And basically, I found there's this new Malayalam movie that came out, which had a song in it, which was very similar to a song from Ratatouille, right? So basically, I was like, oh, same guy for, you know, voice and I basically put down the phone, and I wrote it both and made a mashup and shared it on Facebook.

But because this copyright was owned by someone, they had me and like, it was removed. It was used for copyright purposes, not because they, like, somebody outraged saying, oh, how can you say we have copied and all that? Yeah. So, okay.

So there are community standards in place, right? And basically, they're, and this has been in place for a long time and it's been, they've been calving, bawling with it. So like, one thing, you know, you make up the rules as you go. Okay.

Yeah. Sure. So, to be fair, when they started, I don't think they would have one day become 2.4 billion people. Yeah.

Correct. So, I think the guy started off with, if Zakaba, if somebody told Zakaba that his algorithm for rating women at Harvard would one day end up connecting 2.4 billion people around the world. I think it had started off with a few different principles or something. Correct.

Might have. No, but basically started off with the guy's evil. I just think he's... Yeah.

It basically started with the winkle washes. One, one, the ringle. Right? Winkle Y.

Yeah. Winkle Y. And one, Sakaabag. I'm just saying the next Sakaabag could be hosting simplified.

What? The most interesting part, the most interesting part I had was reading about Zakaba himself. He himself is one of the greatest cases of nominative determinism in some way. Nominative determinism is by the way the concept of people being named according to their last name.

I mean, that profession is being similar to their names and Zakaba literally means sugar mounted. Oh, superbo. Oh, wow. So, it's...

He's basically like a sugar merchant effectively. Oh, he's selling doormine that types. Yeah. So, I mean, what I feel about...

And this is something that you read about how you're talking about maybe the concept of like how morality will like flow through in the 21st and later centuries is that how is ethical thinking or moral code is going to get infused inside tech. And Zakaba again, Facebook is probably one of the first experiments in that you can say it because like exactly what you guys are saying, which was that Zakaba is an evil. He's just... He's a coder.

Right? He's fundamentally sitting there and saying I'm writing code to make my code as efficient and as attractive as possible. I am just chasing efficiency. Right?

He's not really going out and saying that you know what? I want people to like abandon stuff and leave their jobs and like lose themselves over here or kids to commit. So, I'm not wanting any of that. I'm just making my code more efficient.

Everything else is a byproduct of it. And that's why they say that like now the real people who need ethical moral epistemological education in today's DNA are actually the coders. So without that, how will you get like machines in the future who will make like the trolley problem decisions? Yeah.

So, yeah, that's where I feel like the whole thing about these pictures of my research. Mark Zuckerberg who's obviously quite pasty white guy, pictures of him being emotionless and showing him as the next devil. I think it's just all over the top. He's just a coder dude.

Yeah, I would agree. But there's also a lot of black and white, which is what Facebook's community standards are. Right? So basically their goal is they say to earn on the side of giving people a voice while preventing real world harm and ensuring people feel safe.

And they have very effective guidelines as objective as they can be where 18 specific types of content are removed. Right? Including like copyright infringement like I mentioned, but also like you know where visible internal organs and charter burning people are showing etc. And very famously there they have around 30,000 people across the world moderating content.

None of them who want to do it for a living. You know, like they say never read the comments. Imagine your job is for eight hours a day reading the worst and not the worst part of the internet. Yeah.

And there are several reports of that. I think Casey Newton at the Verge has actually done a couple of deep dives into cognizant and Accenture both of which have done some sort of content moderation work for Facebook and or Google. And those stories are just harrowing. Yeah.

I think the most damning one was actually aware that this job will possibly induce PTSD. Correct. And you have to sign a waiver saying that I know that this job is going to give me PTSD and I won't sue Google over it. Correct.

Which I think was wow dude that's that's that's that's that's that's it's not like that. Tag letters don't be evil anymore. But yeah, what I think that these guys interestingly do is say we have content moderators located all over the world. Yeah.

Which is basically proxy for sake wherever we get cheap labor and like. Wow. Wow. So you don't that feature into a month.

Yeah. But so here the interesting case when you started with human moderators, obviously what will happen is human error. Yeah. And they know that when it's a call of a human more than 10% of the time they make a wrong call.

Right. So basically how do you proactively enforce how do they proactively enforce you know policies basically you have to use artificial intelligence for the most repetitive work and a much larger team of humans focused on nuances. Right. Because given that there are 2.4 billion people even an error rate of one in hundred is not good enough.

Right. And then the first one is prioritized certain content. Right. Terrorists and self harm is the first category that was tackled.

And basically 99% of terrorist content is before anyone reports it to Facebook is removed before anyone to Facebook. Wow. So there are a lot of wherever the code works it is enough to do. Right.

Like wherever is black and white it's easy enough to do. But there is a lot of nuance in the rest of it. Yeah. And again, when it's let's say video it's much easier.

So if you just want to remove nudity you can train software to do that. Yeah. And but something like hate speech is extremely subjective or it needs certain nuance. Right.

And that's where the whole idea of, you know, how do we even moderate this comes about because not only does it have a subjective level to it, it has cultural context to it. It is, you know, even abstract concepts like privilege and how that applies and intersectionality. Yeah. Which is why so many because this is such a difficult question to answer which is why some people have asked this better to just shut the damn thing down.

Which I used to laugh at that before. But now I'm like, I mean, how else do you solve these problems? Weekly, one of the best articles I read last year was did the internet just come too early? Was society ready for the internet when it happened?

Yeah. I thought it was fascinating thought experiment. Yeah. But the only way to end all of this is climate change.

It's amazing that climate change has actually become sort of like a don't worry about that. We've done that. But you're a feature. What if Facebook actually did not allow you to post content immediately?

No, that would be terrible. Why? Because that's not free speech then. Right.

There's a great panel to exactly what you're saying Chuck. There was a website that initially began with prevented from prevented people from posting content immediately. And that model failed a lot for quite a while. It's someone decided to screw it.

Let's allow everyone to post. Yeah. Can you guess what website this is? Twitter Wikipedia.

Oh. Wikipedia is actually. Yeah. Yeah.

So it's you try to keep it close. They try to keep it for experts and it didn't work at all. No, it's not. It's very tough.

Yeah. No, no, no. What I'm saying is it's still be open. It's not like all the content is going to operate.

Yeah. No, no. What is it experts? I mean, either approved experts or people whose work will be approved by experts.

Yeah. No, no. That's how we keep it or new media in their case. I'm saying over here to see a lot of, see, it's not going to solve the problem.

But for example, the AI will have a chance to look at it before it gets posted. You know, and once a piece of extremist content goes live, it gets repurposed in many forms before it is potentially taken down. Second, a lot of angry content people after they posted, they may just have a rethink about it and they may decide to, you know, not, I don't know, I mean, it's just a random idea. I'm sure I'm sure I'm going to use the book.

I thought about it. Yeah. But it's completely against speech. Let's say you have delayed speech.

You have visuals of, you know, police violence while the protest is happening. Fair. Fair. Yeah.

Before it gets posted, government gets to intervene and takes a few. Fair. You know what I mean? Yeah.

Yeah. On that rather violent writing. Content. Content.

Content. Contemplative note. Let's take a break and come back and discuss how Facebook is trying to solve this content problem. It feels like everyone's talking about two things right now.

Pickleball and podcast. So what happens when someone who's lived her whole life on the court decides to bring those two worlds together? I'm Sanya Mizra and I'm bringing you a brand new podcast, serving it up with Sanya. I've had some incredible conversations with friends, fellow athletes, artists and entrepreneurs.

Each episode starts with a game of pickleball and once the rallies are done, the real topic is serving it up with Sanya, watch and listen on YouTube and wherever you get your favorite podcast. Are you looking for finance products and services that can enhance your personal finance experience? Are you looking for a space to talk about your financial product or service? And are you looking for a Chris Rock Show where the conversation is all about money?

Well, your search ends here. Hi. My name is Arupak Gupta and I'm host of the best of our podcast and I invite you for the conversation about your personal finance for each Monday on the IVF podcast app or the website or not any podcast streaming platforms. See you folks.

Hi and welcome back to the Tony Sebastian podcast. We're now taking a look back to our host. This is a good segue into the next thing that we want to talk about which is fake accounts and fake news on Facebook and how they are actually tackling that problem. So one of the great things again about Facebook which a lot of people have a problem with is their insistence on usage of your real name.

Which basically means that they have removed over 1.5 billion fake accounts just by virtue of this. And also the integrated nature of Facebook, Instagram and WhatsApp means that if you have created a fake account on one platform they're able to remove fake accounts across the world. You know one set of people who faces a problem with this is people whose surname is actually is a word. Like I have a friend whose surname is contractor and she's unable to create an account on Facebook because of this.

So what she's done is she's had to change that C to that friend Chwala C, Garcon. She's had to change it to that. So everybody why is your name is contractor? So I'm guessing Faroq engineer has the same issue.

He's not a bad word. Go by the question as far as I know is not a profession. I mean, Thade I'm assuming as many of my classmates in a good chat constantly told me is a auditory. So a auditory is a auditory.

So a auditory is a auditory. So a auditory is a auditory. A village of auditory. A village of auditory.

A village of auditory. How have you never gone there? A village of auditory. To this fact.

Anyway. But you know how Trevor Noah talks about basically my cultural context to Africa is Trevor Noah. How they have a lot of names like table, chair, etc. They'll have a nightmare creating accounts.

I think they have bigger problems than having a conversation. Okay. I mean 2.4 billion people use it on Facebook. They might be wondering whether the remaining people are well there.

Okay. So basically they use both the technology and human review to remove fake accounts. And basically there are different types of fake news. You know, just information.

Other can be led to economic incentives. Like you create a fake thing and then you run a phishing scam to get people's money. So that's easier to shut down. So you're so charmingly old school.

That's that's that's that's. I mean, I would love to go back to a time where the first thing that you could do on the internet was scam somebody for their money. It seems like such a charm. I would take that.

It's the classic. I would go back to Congress taking away all my money to what is happening right now. But I still see those. One of those things that I keep seeing is by these ribbons and glasses for one rupee.

I don't know that post keeps coming up a lot all the time. But now we have like UPI scams. It's it's it's it's it's it's off Facebook. Yeah.

There's a network show called Jamdara, which is all about this place in Jamdara, which is where pretty much the epicenter of all the UPI based phones scams that people have. So yeah, it's it's pretty awesome series. I mean, they that place actually at one point in time, like for a year, they drew in about 50 lakhs worth of money or some stuff like that. Oh, where the UPI center?

Oh my God. Oh my God. Yes. UPI center.

UPI is. Okay. No, wait a minute. Visual.

I love how you have a whole ecosystem. Yeah. Yeah. You basically like suck his Facebook Instagram for itself.

I have the thoughts. Oh, no, please let that not be. Okay. So basically wherever there's a financial incentive, they've tried to remove that and it's easier.

And in certain countries, they also work with third party fact checkers who are all technically nonpartisan and international fact checking network has to certify them to be able to log in. Right. So the major problem with tackling fake news is the problem of scale. Like you can create fake news faster than you can fact check it.

Yeah. So it's not been as successful as it can be. But also, I think as of this week, a couple of days ago, they even signed up Reuters because the problem of fake news is getting even worse with deep fake videos and stuff like that. I think there's a famous video of Nancy Pelosi and yeah.

Yeah. Yeah. All of that. Yeah.

Although on that note, although I'm not a fan of what's happening in the world right now, it's kind of like forcing tech to, you know, solve these problems like last week or something. Google released a feature for journalists that will allow them to detect deep fake videos by putting together all the common, you know, deep fake techniques and figuring out background difference. You can figure out, okay, this person has been removed from here or whatever. It's a really sad tool.

I think it's a good thing. I don't think a tool like that would have come across if, you know, shit didn't happen. I'm not saying that shit should have happened. But what I'm saying is like, it's all this is sort of like, you know, if you think of the world as a body, we are currently going through a disease where the immune system is now.

Okay. You can figure out what's going on. This is left as an exercise. You can write in your best answer.

You can chat about the Gmail.com. We will create that email id and the best answer wins dinner with Naren. Okay. So that's a good cook, by the way.

Yeah. Very. In other cooking stuff like fake news, basically fake news, I mean, fake news, the problem is that you can only reduce the distribution of it, right? Because once you have a free speech, anyone can put anything up.

But with the help of these third party factors and their own facts, they're able to identify false news, show false stories lower in the news feed, provide more information if you see false news. So they'll say, you know, provide context. Also, in some cases, you know, just have code taken out of context. Yeah.

All of that. And also they want to notify people if when they have shared false news to tell them that, you know, what you've shared is a false piece of information. I actually saw one of those. Like, I've seen this news that comes from this source is typically false.

So please, yeah, I've seen one. You see that this thing, right? Uncover this as graphic information. You can uncover that thing.

Same thing with fake news as well. I think you should stop falling. So I just want to say that you should stop falling. So I just want to say that.

Once. Once. Once. One article.

Yeah. But again, like we discussed earlier, what won't be factured is, of course, political ads. And I have my own other personal pet peeve, right? Which is that people are not necessarily knowledgeable about satire or parody.

And I think there's an easy problem to fix, right? Because you have like a fake news or whatever, which they can tag as parody, which they've not done. Yeah. I mean, when your country's prime minister is quoting fake news, or I don't know if you guys saw it, Trump recently tweeted out a curb, your enthusiasm clip.

Basically, it's making fun of Trump supporters. But he basically tweeted out saying tough guys for Trump, right? Oh, yeah. Because I remember this.

Yeah. His followers don't get nuance or parody or anything. So it's good content for him. Which is my Tony.

You and I are still alive. Oh, my God. Oh, my God. Oh, my God.

Oh, my God. Oh, my God. The other thing that they content framework does already, community standards is that borderline content, when it reaches a certain scale, they'll throttle it. Which is why we have our favorite humans of Hindutva, which is a lovely page, but it keeps complaining from time to time that his reach is being throttled, right?

Because it's an active step by Facebook because they call it borderline content, right? And they say at scale, it can undermine the quality of public discourse and lead to borderization. So in your opinion, or at least from what you're read, it's not necessarily that is because the government is complaining about that page, or could that be a factor? It's a bit of both, right?

One is that once it's been flagged, it'll be evaluated to see whether it is. Sure. But if any of our post is removed, it is because someone has reported it and has not met community standards. But there's an additional guideline that they have, which is to say if it is borderline content, then you reduce the distribution after a point in time, right?

Not like community standards. Like community, but beyond that. And then comes basically the impossible question of algorithmic fairness, right? Which is basically at the intersection of ethics and artificial intelligence.

And basically in philosophy itself, people are not able to agree on how to define fairness, right? So for instance, there are two schools of thought. One talks about equality of treatment, and another talks about equality of impact. And these are not even philosophical problems that can be solved.

So how does an algorithm add up? Sure, yeah. For instance, I like privilege that we spoke about, right? Like, you know, who has more privilege over whom and therefore does that become a criteria for fairness and cultural context?

All of this. But basically what that means is that currently there is limit to how you can effectively moderate content, even there are 2.4 billion people on Facebook. Right? So what they have currently is an appeals process and basically artificial intelligence and human review, right?

But what they want to create, and this is basically what is interesting and everyone's looking forward to, is what is informally called the content supreme called. Oh, yay. Is that what it's going to be called? So it's called oversight board, right?

Yeah, it's called Facebook oversight board. It's actually my side. Yeah. And the whole idea behind this is that the Facebook shouldn't be making these decisions, right?

Because it had to, because so far they had no choice, but now they're like, we need to find out a better way to do this. And three key aspects is to prevent concentration of too much decision making within Facebook, create accountability and oversight, and provide an assurance to everyone that these decisions are made in the best interest of community, and not because they want to make a shit-ton of money. It's not why this content exists on Facebook, right? So then basically this means you create an independent body, which like you said, is called the oversight board and not a Supreme Court.

Independent of what? Independent of Facebook or Independent of, like what I'm asking is this going to be, I know that these guys are going to be funded by Facebook, or the salaries, whatever it would be paid by Facebook. But are they part of the Facebook organization's structure? Are they going to be Facebook employees?

Yeah. So essentially, thank you for that question. But basically there's a nice rock paper says going on. How?

Spock. Yeah. What is called the oversight board, which is the board that is the Supreme Court for content. Then there is a trust, and then there is Facebook.

So the trust basically, Facebook funds the trust. The trust ensures the government and accountability of the board and the board's adherence to the stated purpose. And the board, the oversight board, what it can do is it will review content which has been reported to it. And it will take certain binding decisions, which then Facebook has to comply with.

I'll comply with it. But it can also advise Facebook on content policy, which they can be like, oh, thank you very much. Thank you. They can ignore it.

Or they can act on it. So basically, or another way of asking what I asked actually was suppose the oversight board grows big enough. Will they also reach a stage, for example, where they started advising Twitter, YouTube? Yeah.

But this currently is only in the ammeter. Yeah. But it could evolve from there possibly. Yeah.

And independent of company. That's really what I meant. Yeah. Even right now it is because they've, Facebook has already pledged 130 million to the trust for six years, which I think was a much larger amount than people.

You're literally buying trust. Yeah. Absolutely. So currently in the present scheme of things in the first half of 2020 is when it's supposed to come about.

The size of the board is 11 to 40 members. Right. And members must have, you know, they are supposed to come from diverse sets of backgrounds and should have familiarity with digital content, governance, free expressions, civic discourse, safety, privacy, and technology. And you have six years experience.

Yeah. I'm not going to be able to get to the other side. So are they all going to be physically located in one place or spread around the world? I think they are.

I'm not very sure about that. But basically. I think they're all going to be in one place to begin with. Yeah.

And another sort of second balance is to say that each number will serve for a three-year term and a maximum of three terms. So that's a number of situations where someone is running. Nine years maximum. And it's a staggered sort of term so that there is always continuity in the, you can't sort of stack the board.

I mean, you can, but basically it is supposed to exist independently. And like one of the things that Facebook has done this proactively, and it's a good thing is because the other mechanism to do it would have been for governments to lay down. Yeah. Which is always not going to work out the way you want to.

Or at least not how Facebook wants to work out. Yeah. Or the common people. I mean, basically the state as it is as too many powers, you know, and arbitrary laws, etc.

So start messing with the content that you consume, then it's going to be a really problematic process. So right now, what will they review? Because it's only 11 to 40 member team. Basically, even the appeals process and community standards continue to be for content that is removed.

Only for content that is removed, you have an opportunity to appeal to the board. Okay. Okay. Now the users can appeal or Facebook can itself appeal.

And then it is brought before the board and they review certain content types. But isn't that process in itself flawed because there's a lot of content that you'd want to, like, that is there that people object to, like for example, that is some really controversial. Hosts by, I mean, I know political content is example, but if there is something like that, that should potentially have jurisdiction within that content supreme court, right? Yeah.

So right now there's scope of execution is only for content that has been removed and, you know, sort of directed to it. And they've basically said, I mean, you realize that, I mean, it's only like 11 to 40 people. So there is a, it's not like they can review every damn thing. So it has to be like, so they've sort of basically put it as, look at indicators that demonstrate specific cases that are specific cases that are specific.

And difficult. So significant means that, you know, the content has real world impact and is of a high severity, large scale is affecting public discourse. And in terms of difficulty, it is that, you know, it can be disputed that there's a disagreement about Facebook's decision. It can also be uncertain or it can be competing because again, like you, one person cannot make a fair judgment on what is what is correct and what is not.

Right. So basically that is the ambit of the Supreme Court and it is supposed to come, I mean, the oversight board and it is supposed to come into effect in the first half of 2020, which is before the elections in the US that that is the basic process. Right. And on its own, Facebook can also expedite certain content questions.

And basically a decision will be arrived at within seven days. But if the oversight board's decision violates local law, then local law will take precedence over whatever. Right. Yeah.

And basically the board cannot take longer than 30 days to come to a decision. Right. And if there is a second round of appeal, it has to be done in 90 days. Wow.

So yeah, basically that's what the oversight board is. So in a sense, I feel that, you know, basically it's unacceptable to have Facebook making all the decisions by itself, but also like government making those decisions is much, much worse. Right. So it's like a compromise of sorts.

Yeah. This is like a compromise of sorts. It is a step in the right direction. Yeah.

And they are, you know, promising transparency and accountability. And the sheer scale of Facebook means that there will be enough tech writers and policy writers criticizing them anyway, which means that there is a scope for them to grow up and improve and stuff. So what I put what I like about this whole thing is like, and I don't know whether it's going to work or not, but I think that's a good, that's the case for any experiment. Right.

Like you never know whether something's going to work or not. What I like is genuinely different from all the other solutions that are out there in the market, which are just either higher, more moderators or AI and machine learning will catch up or implementing more of AI. And this is a totally different sort of, I don't think it's going to reduce the necessary, the requirement for AI and ML to help solve this problem as well, but it's a genuinely different sort of model altogether. And it's really worth seeing where this goes in a couple of years.

I think he has plans of scaling that number, which are what you mentioned that 40 or 11 to 40. I think he also scaled up to 3000 or something in a couple of years or something if all goes right. Yeah. By interesting, if that become takes off as a different career by itself and law colleges now start teaching like, yeah, content judge, content judge, content judge.

It could be a thing, man, like two million a back if they said if two people fight, then they will be a person who's professionally paid to resolve it. People are loved. Yeah. People are fighting ads that needs to be some fellow who's actually waiting PPD and saying bullshit.

Okay. But yeah, just as we speak, I think a couple of weeks ago, TechCrunch has already called Facebook's oversight board week and toothless. So yeah, this is not a debate or a discussion that will ever end. But they've hired the first person from a nonprofit freedom of expression and digital rights group, who will, you know, in the administrative capacity put together the rest of this.

So from his from Suckerburg's draft version in 2018 to them putting out a charter and now actually getting people. Which is actually fairly quick for something like this. Yeah. Not for a company that had a tagline that said move fast and break things, but I think we are okay with moving slowly.

Yeah. I think the breaking things are democracy. Yeah. I think it's a good step.

I look forward to how it's going to go. Yeah. One thought experiment then suppose Hillary Clinton had one 2016 and misinformation and all that didn't happen to the scale, at least in the US. I think India's story would have been totally different.

I don't know how related what's happening over here is to what has happened over there. I think that's a bit of a stretch. How different do you think things would have turned in terms of fake news and all that, especially since a lot of what we're talking about is US centric. Would we have ended up with a oversight board?

Yeah. It's a it's a tough question, right? But I'm not sure that I mean politics obviously has brought it to the forefront. But like I said that Wikipedia article already has like, you know, 100 different criticisms.

So eventually the scale problem would have to act in, you know, it's just content moderation overall, not necessarily just fake news. I think it's a scale of Facebook rather than the turn of events that are disturbing. And they want to be profitable and last forever, right? So it's in their interest to get their act together before some government decides to shut down.

Yeah. And I think this is something like, I mean, this is similar to the thought experiments people have about if a certain thing had not been invented kind of thing. But I feel that there have, I mean, fake news. I mean, I think what was much more, if we have seen the Cambridge Analytica documentary, what was much more deep rooted as far as fake news was concerned and much more widespread impact happened because if it was the exit movement, right?

So I feel that maybe that would have been the epicenter and therefore the driving force. I mean, of course you can hypothesize whatever this might have ended up. But I feel that, yeah, one way or the other, we would have ended up over here. Maybe not in its form that it is, but as an Ombudsman body, whether it's people oriented and later probably tech oriented, I think this stage is probably inevitable.

And it's probably the beginning of a lot of content Supreme Court's or legislative bodies about what is right and what is not about how the internet works and how people are connected now. The first time the word Ombudsman has been used in the fact that it's like exactly what he said. So is this board can either evolve to be a Facebook only board doing only Facebook work as it were or just taking off from what you said it could be different. I see it getting inspired from this tomorrow for instance and starting their own.

Yeah, I mean, I don't think anyone... It'll be a model. That's another reason why this is important. So yeah, I mean, as much as I'm on the side of Facebook, I don't think Sakaabak wants to do this.

He was forced to do this. No, no, no, no, no. I don't think anyone is in the inspiration. I don't think anyone is going to...