Every prediction from Balaji's podcast with Tim Ferriss
Plus additional commentary from Balaji
Balaji’s website describes him as:
[…] an angel investor and entrepreneur. Formerly the CTO of Coinbase and General Partner at Andreessen Horowitz, he was also the cofounder of Earn.com (acquired by Coinbase), Counsyl (acquired by Myriad), Teleport (acquired by Topia), and Coin Center.
He was named to the MIT TR35, won a Wall Street Journal Innovation Award and holds a BS/MS/PhD in Electrical Engineering and an MS in Chemical Engineering, all from Stanford University. Dr. Srinivasan also teaches the occasional class at Stanford, including an online MOOC in 2013 which reached 250,000+ students worldwide.
People who follow Balaji know him as the person who correctly predicted the severity of the coronavirus pandemic as far back as January:
Inside of certain circles, Balaji has a reputation for prescience. Conor White-Sullivan (founder of Roam Research) said it well:
If you’re unfamiliar with Balaji’s work, he recently started a new project at 1729.com, a website that pays you cryptocurrency to complete micro-tasks. This is part of a broader effort to bootstrap something he calls “The Network State.”
Balaji recently appeared on the Tim Ferriss podcast to discuss a huge range of topics, including cryptocurrency, the future of politics, problems with the current state of journalism, and more.
Throughout the podcast, Balaji made a number of predictions, some with explicit and falsifiable claims. In the spirit of intellectual progress, I’ve collected those predictions here so we can revisit them in the future.
This is not unusual; blogs like Slate Star Codex and Nintil routinely post public predictions and bets. If you’re actually seeking the truth, being able to identify where you were wrong is a great way to calibrate your mental model.
Balaji has been compared to Cassandra, a figure from Greek mythology who was granted the gift of prophecy. Unfortunately, Cassandra was also cursed so that no one would believe her. Balaji was our Cassandra of 2020 — let’s try not to let that happen again.
Method
I listened to the entire podcast and took notes whenever Balaji mentioned a prediction with a specific date or resolution criteria. Those predictions are listed below under the Predictions section. For topics that were interesting but lacking a specific falsifiable prediction, I listed the quote underneath the Quotes section.
I also reached out to Balaji to see if he had any additional input on the predictions offered in the podcast. While he did offer comments, he noted that they weren’t extensive. I’ve added his comments below, verbatim, with the prefix [BS].
Okay, here we go.
Predictions
For example, one thing I anticipate will happen sometime in the 2020s, I think in some country, probably the U.S., is that the cloud will burst. […] One of the consequences is that it shows it’s not a theoretical risk to think about what these hacks can be. […] All Facebook DMs will rain down. All Slack DMs will rain down. [These data] will be available and searchable, Wikileaks style.
[BS] These are just examples of what I mean by the cloud bursting. Arguably this has already happened with the 500M FB and LinkedIn hacks. Also this is actually a book reference — I can’t remember the name but there’s a book that mentioned this concept, in similar terms regarding the cloud bursting
[2030] Maximize a $100k investment
So the question is, if I was given $100,000 […] what do I do with that? Like, how do I maximize returns? So the dumbest thing, but that, I think is the most obvious thing to do is put half into Bitcoin and half into Ethereum. [That is] the simplest thing that I think is, you know, guaranteed to produce really exceptional returns. In retrospect, from 2030 looking back, there’ll almost certainly be something you could do that would be better than that, but that’s going to do extremely well.
[BS] What follows below is not investment advice, just my personal opinion and letter to myself basically.
In the podcast I was a bit stronger — i said “there will very likely be something better than that” — because there will be investments that appreciate more over the same time frame — but for the scale of money invested this strategy it would do very well.
Also my recollection is that Tim asked about investing $100M as well as $100k. The strategy for $100M is much harder than $100K as you can’t multiply large amounts of money easily whereas $100K can more easily be multiplied 1000X via good seed investing.
So first let’s do the $100k. If you are an angel investor you can get great seed deals, with a shot at a 1000-10000X investment. So that is on the table and may outperform the 50% BTC/ETH strategy.
But it requires time and effort to find those deals, and the average Joe doesn’t have access to Facebook as a private company in 2005. They do however have access to BTC and ETH via crypto exchanges. And those assets will likely do better over the next 10 years than many public stocks, especially if you hold the keys locally so you can — at your option — benefit from every fork. So even putting $100k into BTC/ETH is not a bad strategy for someone without deal flow.
As for $100M, you can’t put that into a small company. But you can put it into BTC/ETH without moving the price much. They are huge assets now. As before there will again likely be growth stage tech investments or other coins that outperform the $100M into 50% BTC/ETH strategy. But for the level of simplicity and scale this offers, it’s a pretty solid bet to multiply large amounts of money over ten years.
[2030] India as a media superpower
The other aspect of crypto is it would allow [for] Indian media […] You want dark horse prediction for the 2020s? I actually think India has the potential to become a media superpower.
[2030] Zoom filters for pseudonymity
What I think is going to happen soon, sooner than we think, probably 2025 or 2030 latest, is you'll have filters on your Zoom that are not simply a background to San Francisco or whatever, but audio and video filters so that you sound and look like a different person.
[BS] This feature may end up being in another app other than Zoom. It may even be in VR rather than Zoom. But people will use their real faces and voices less in video chat by 2030. Partly for pseudonymity, and partly because it’s easier than using makeup, and partly because it just gives total control over how you are seen. Snapchat filters are just v1 of this.
[2031] China network-level attack on Bitcoin
The attack on BTC that I think would be the most likely is that the Chinese put up the great firewall against it. So it's a network level attack as opposed to a mining attack. [...] That would be the worst case scenario. Remember two things that have to happen there: first, they have to put up the firewall, and secondly, they have to allow Chinese mining to continue. [...] I'd say 30% chance of a firewall attack. It's not 100%.
[BS] One point I should have made more clear is that I think they may put up the firewall, but it is somewhat unlikely for mining to still proceed despite the firewall. The best-case scenario if the firewall goes up is that Chinese miners all shut down domestically too— and move operations abroad— rather than stochastically mining which would cause the peekaboo problem I mentioned. I don’t think it’ll be possible to shut cryptocurrencies down globally though given proof-of-stake, progPoW, and other mining algorithms.
Quotes
The journalist and the subject
The contrast to this is like a journalist versus the subject. Podcasting is peer to peer. Journalist/subject is hierarchical. There's a journalist, a corporate journalist, an employee of immediate corporation. And then there's the subject, as in, you know, the subject under the microscope or the subject and there's a king and their subject. And that word, the subject to. That word subject to is such a revealing word. It gives away so much: journalists and the subject under a microscope. And the thing is a journalist / subject relationship. When the journalist interviews the subject, journalists can edit, they can pull whatever quote they want, and they have the distribution that the subject doesn't.
One of the biggest things that [journalists] agreed on is journalism's biggest blind spot. . “Groupthink. We draw from a limited pool of people who generally have a similar background class, are simply unable to see the perspective of people who are not like them and tend to drive out those who don't fit in.”
Now I actually understand this. Coming from academia, I could see how that episodic bubble happens. It's ironic that the people who complain often most about filter bubbles are actually complaining about the plural because there was just one filter bubble that they controlled, and now they're annoyed that there's more than one.
[BS] Quote from NYMag article from 2016: http://nymag.com/daily/intelligencer/2016/07/case-against-media.html
Financial independence as ideological independence
You should radically reduce your expenses if possible, because financial independence is upstream of individual independence and ideological independence.
Negative press as an attack vector
Negative press is an attack on your social network. How do you mitigate this? Well, first, as I mentioned you are you're like a quote sovereign individual, you’re financially independent. So if people pull away, it doesn't matter. But second and more important, you need to actually identify where likely attacks are going to come from and make sure you actually choose your friends, your employees, everything such that that's actually robust to foreseeable kinds of attacks.
[BS] Later I talk about collective independence too.
Asymmetric weaknesses to negative press
One interesting consequence of this, if you flip it around, is if you have a monochromatic supply chain. There's basically nobody, for example, at The New York Times Company who reads and respects Breitbart. So Breitbart has zero effect on the New York Times company. But if you go to Breitbart’s extended social network, to their supply chain, there's absolutely folks you know, whether it's people running ads or other kinds of things that still respect the Times. So [Breitbart] can not publish an article that has effect on [The Times], but [The Times can publish [an article] that has an effect on [Breitbart]. And by that I'm not saying anything pro or con about Breitbart or the Times — just making an observation of fact.
Another example is in China, the Chinese People's Daily. If you're Chinese and they denounce you, your whole social networksupply chain is like, “Whoa, the government just denounced this guy. You're in trouble.” If you're in the US, nobody even reads Chinese or even cares.
Real names are actually a technology. In fact, real names are a stolen base. In many ways, even the term real names is stolen base. It's better called a state name or a Social Security name, because [it’s a] global identifier. It's an identifier that anybody can type into a database and pull up all kinds of information on you. And this is something where historically past cultures understood that giving out someone's real name was actually very bad or knowing it, you had power over that person.
On the difference between pseudonymous and anonymous names
So the point being that the alternative to real names, or like a single global identifier and everything, is pseudonyms, search resistant identities, which people are already using at large scale. There's 400 million pseudonyms or pseudonymous users on Reddit. And by the way, there's a distinction between anonymous, pseudonymous, and real name. Anonymous is totally disposable identifier. That's like the culture of 4Chan. Pseudonymous like Reddit is a persistent pseudonym that accumulates reputation.
On the popularity of pseudonyms
You know, I think it's gonna become less and less and less common for people to give out the mapping between their physical body and their user names online. Like you may not even tell your wife, you know. [You're not going to correlate those two things for someone.] It would be considered a faux pas.
If you think about the root of democracy, it's not actually voting per se. The democracy is the consent of the governed. That's what makes a democracy legitimate.
Types of votes in the U.S. political system
There are actually three kinds of votes. We think of two kinds: voting with ballots and voting with dollars, which are democracy and capitalism respectively. There's a third kind, which is voting with feet and the migration. Actually all three are components of the U. S. Political system.
Media -> digital, finance -> crypto
Actually, the two biggest stories of Internet in India over the last 30 years are the economic liberalization and the Internet. Crypto combines both. It is the financial Internet. It's literally at that level because what did the Internet do? It digitized books, movies, songs, all forms of media, right? Crypto goes and digitizes everything scarce: stocks, bonds, commodities and of course, gold and other things. The internet was the way to copy things from point A to point B. Blockchain is the way to transfer things, scarce things, in a proven way. So every financial asset becomes natively crypto, just like every media thing became natively digital.
Ideological mapping independent of national identity
States no longer map to each other because people are friends with millions of people around the world who share their values. But they're not physically concentrated. The physical areas have less volition by their subjects, so they have to enact more coercion and more coercion in turn leads to less volition and less volition leads to less legitimacy. And that's a negative feedback loop where at the same time people are building bonds with other folks of like mind that are geographically dispersed. And that is something that's exerting a strain on the whole system. We should think eventually this results in mass migrations and people assembling into communities of, like mind to sort of rebuild civil society with folks who they agree with ideologically.
Turning a constant into a variable
There's a loss of faith in the past, and that makes people open to turning a constant into a variable. [Consider] a dollar sign. [...] Something that was a constant is now a variable. Fiat or crypto.