Balaji Srinivasan: The Man Who Called COVID

At first, Balaji Srinivasan held his tongue about the coronavirus. 

Years before he became an angel investor and intellectual firebrand of the cryptocurrency community, Srinivasan had co-founded a genetic testing company that was eventually sold for $375 million. He had also written papers on microbial and clinical genomics and taught bioinformatics at Stanford. So even though his degrees were in electrical and chemical engineering, he knew a thing or two about biology. 

And what he was reading in January 2020 about the novel virus, in the scientific literature and the signals coming out of China, gave him pause.

“I was like, this looks actually very serious,” Srinivasan recalled in an interview on Nov. 26. “And it is not being covered enough. And it’s being treated as if it’s ‘something that happens to foreigners, haha.’”

Nevertheless, he said, “I actually hesitated to tweet about it for a while, simply because it was going to be seen as, you know, ‘paranoid’ or whatever.” 

As the media and authorities in the U.S. continued to downplay the danger – “Don’t worry about the coronavirus, worry about the flu,” BuzzFeed News reassured its readers – Srinivasan felt compelled to speak up. 

“Normally, when you trust this stuff,” he said of news outlets, “you just lose money. Here, you literally could have lost your life.”

On Jan. 30, a week after the Chinese government imposed a lockdown on the city of Wuhan and two days after Buzzfeed published its later substantially rewritten article, Srinivasan shared his thoughts with his followers, who numbered over 130,000 then and are nearly double that now. 

“What if this coronavirus is the pandemic that public health people have been warning about for years?” he tweeted in the first of several threads that turned out to be tragically prescient about the threat. 

His concerns about being viewed as a Chicken Little were vindicated a few weeks later by an article in Recode (more on that, and his public confrontations with the tech news site, shortly). But that perception was obviously short-lived because the coronavirus went on to claim 1.5 million lives worldwide, shrink global GDP by 5% and exact an incalculable psychological toll of isolation, boredom and mask-wearing, elbow-bumping, six-foot-apart misery.

Srinivasan was spot on about the virus. He has a point about the media. More broadly, he’s emerged as a prolific and perspicacious thinker about the erosion of trust in experts and institutions, the breakdown of consensus about truth and how these might be rebuilt. These are some of the core problems that Bitcoin or its various offshoots and mutations (loosely grouped under the rubric of “blockchain technology”) attempt to solve. 

“The thing that stands out about Balaji is he’s incredibly well read about a lot of topics,” said Preethi Kasireddy, who worked with Srinivasan in the mid-2010s when they were both partners at the venture capital firm Andreessen Horowitz (a16z). “And so he’s able to combine his knowledge of historical events and crypto and math and he’s surprisingly good at understanding people and their motivations as well. When you’re able to do that, you can think at a level above most people.”

Besides his prognostications about the virus, Srinivasan’s big ideas about the future have made him one of the most influential people in crypto in 2020. 

Before we get to those big ideas, allow me to break the fourth wall and preemptively address some potential criticisms of this article. If self-referential journalism annoys you, skip to the next section.

Some readers may wonder why CoinDesk is recognizing Srinivasan now when he hasn’t had a full-time job since leaving crypto exchange Coinbase, where he was the chief technology officer (CTO) from April 2018 to May 2019. This would miss the point. 

“The irony is he stepped down from Coinbase and actually very little of his work this year was crypto-focused, per se. But I think he hit on a few big macro trends that are all going to be important areas for crypto to upgrade or replace in the years to come,” said Ryan Selkis, co-founder of crypto market data provider Messari, who does not hesitate to call Srinivasan a “visionary.” 

Another foreseeable objection is that Srinivasan’s crypto projects have “failed.” His first startup in the sector, 21.co, pivoted from bitcoin micropayments hardware to paid email and tasks under the name Earn.com before it was sold to Coinbase for an undisclosed price in 2018. Nakamoto.com, the highbrow crypto journal he launched in 2019, hasn’t published since January. 

See also: Ex-Coinbase CTO Is Behind Mysterious Nakamoto.com, Sources Say

Not that all his crypto business ideas have been flops. While paid email hasn’t taken off, Earn’s final iteration under his leadership, rolled out in late 2018 after the acquisition by Coinbase, appears to be a hit. Known as Coinbase Earn, this service pays users small amounts of cryptocurrencies for taking courses about them. It’s prominently featured on Coinbase’s homepage and mobile app. CoinMarketCap, the popular data site owned by the Binance exchange, copied the idea, as did trading platform CoinList. 

“The acquisition of Earn opened up a number of opportunities for Coinbase,” said Emilie Choi, the exchange’s chief operating officer. The service allows users “to do more with their crypto than just buy, sell or hold – it encourages them to really explore crypto’s utility in a way that’s fun and approachable. Asset issuers have seen massive value in the combination of Earn and Coinbase, committing hundreds of millions of dollars to Coinbase to bring mass-market consumers into crypto.”

Among those issuers was the Stellar Development Foundation, which committed 1 billion XLM (worth about $170,000 at recent prices) to Coinbase Earn last year.

Srinivasan also made an impact on crypto during his time at a16z. Kasireddy credits him with “seeding the idea” of crypto’s potential with co-founder Marc Andreessen and general partner Chris Dixon. 

“Obviously, Marc and Chris are smart enough to understand why crypto is important, but he was a huge influence in the firm’s desire to go deeper into crypto,” said Kasireddy, who later founded a social network called TruStory. 

One more complaint is likely to come from my colleagues in the Fourth Estate: Why write a flattering profile of someone known for bullying members of the press? 

The answer is… Well, it’s complicated. This brings us to the Recode incident. 

On Feb. 7, Recode’s reporter wrote to Srinivasan requesting an interview for an upcoming story on “concerns about the coronavirus in Silicon Valley/the Bay Area.” Instead of writing back (“thanks for the opportunity but I’ll let my tweets speak for themselves” would have been a standard response, or even a curt “no comment”), he called her out on Twitter.

Above a screenshot of the journalist’s direct message, Srinivasan typed:

Media people (myself included, at the time) found Srinivasan’s behavior inexplicably belligerent. Contacting a subject of an article for comment is a basic journalistic duty. Who would attack someone for doing her job? 

In subsequent tweets, Sirnivasan explained he did not believe Recode was acting in good faith. Reading between the lines of the reporter’s message, he guessed the story would be a “soap opera piece” about venture capital firms adopting policies against handshakes.

As it turned out, he guessed correctly, right down to the headline, which referred to a sign on the door at a16z: “‘No handshakes, please’: The tech industry is terrified of the coronavirus.” 

Recode’s Feb. 13 article suggested that some VCs had “gone too far” in their cautionary measures and comments, quoting one observer who called rich technologists’ fear of germs “paranoid.” It contrasted Srinivasan’s tweets urging San Francisco to cancel its Chinese New Year parade (“a beloved tradition,” the article noted) with sanguine statements about the risk posed by the virus from local authorities, including Mayor London Breed, who allowed the parade to go on as planned Feb. 8. 

Srinivasan responded with a 20-tweet thread disputing the Recode story’s claims, one by one. 

A month after its publication, San Francisco was put on lockdown. 

In November, the local Chamber of Commerce canceled the Chinese New Year Parade for 2021.

In retrospect, the Recode piece aged poorly, quickly. (Its author did not answer a request for comment.) But until our interview last month, I still couldn’t fathom why Srinivasan had responded so rudely to Recode’s initial inquiry. 

Explaining why he took straight to Twitter, he used a streetwise analogy. 

“If somebody’s telegraphing that they’re going to punch you, and they’re winding up like this,” Srinivasan said on our video call, pulling his fist behind his shoulder like a boxer, “it’s pretty dumb to be like, ‘okay, I’ll wait for the punch, and then spit out my teeth.’ No. Your answer is, either get out of the way, block the punch, or [shout] ‘hey, this guy’s trying to punch me’ and you point to the crowd.”

Even if the story was conceived from the outset as a hit piece, I countered, wouldn’t getting on the phone with the reporter increase the odds, however marginally, of a more balanced article? Surely refusing to engage meant forfeiting the opportunity to shape the outcome. 

Srinivasan replied that engaging would have been a reasonable strategy prior to 2013, the year he says the so-called techlash began.    

Since then, the custom of most media companies has been “to treat tech companies as the enemy,” he said. “All the prizes, all the culture, is built on dunking on the tech bros.” 

Lest you think Srinivasan considers Silicon Valley beyond criticism, he places some of the blame for this problem on tech itself, not least of all the platform that serves as his main megaphone.

The old-school, pre-2013 reporter “wasn’t as incentivized by Twitter to just demonize [individuals and companies] for clicks, because what Twitter does, it trains you to strip context, add exclamation marks,” he went on. The platform’s coveted “blue check,” created to attest that a user was who he or she claimed to be and not an impostor, became “a mark of prestige,” which in turn “led to this feedback loop where to gain status was to be very negative, polarizing, etc.” 

Simply put, in his view there is nothing to be gained from talking to someone whose success is measured in scalps. 

To be fair, adversarialism has been a press tradition going back as least as far as H.L. Mencken’s time. It’s not always sadistic; often it’s salutary, even when aimed at tech companies. The Wall Street Journal’s investigative reporting on the alleged fraud at Theranos comes to mind.

Yet, Srinivasan’s description of the state of my profession has more than a grain of truth to it. And tech is not the only punching bag. Obscure, harmless individuals who express dubious views on social media or commit inconsequential gaffes at parties are regular targets for dunking. 

(CoinDesk, I will readily admit, has made its share of blunders. At least these were due to haste, not hate.) 

The media struggles with a misplaced sense of epistemic certainty. Srinivasan phrased it less delicately: “Absolute arrogance and lack of humility, frankly, by people who don’t know and don’t know that they don’t know.”

Which brings us to the interesting part of the story.

Recall that early on in this crisis, health authorities advised the public not to wear masks. 

“Seriously people- STOP BUYING MASKS! They are NOT effective in preventing general public from catching #Coronavirus,” tweeted the U.S. Surgeon General on Feb. 29. Only on April 3 did the Centers for Disease Control and Prevention (CDC) recommend all Americans wear masks in public, changing its previous stance that only the sick and their caregivers needed them.

See also: The Decentralized Web Has Plans, if Not Solutions, for the Misinformation Nightmare

“There was so much official misinformation flying around,” Srinivasan said in November, describing the environment of the first two months of 2020. He rattled off some examples: 

“‘The flu is more serious.’ ‘Travel bans are overreacting.’ ‘Only Wuhan visitors are at risk.’ ‘Avoiding handshakes is paranoid.’ ‘The virus is contained.’ ‘Tests are available.’ ‘Masks don’t help.’” 

Most if not all of these memes were widely discredited by the end of March, he noted.

Soon the official narrative took a 180-degree turn. “People were fighting masks for actually quite a while. And, then they just completely flipped as if they had never fought them,” Sirnivasan said. 

The failure of leadership is one reason why he can’t fully blame the people who refuse to wear masks and scoff at the risk, even though he considers such decisions irrational. 

When politicians throw lavish dinner parties or visit their hairdressers during lockdowns, “it shows [the public] that the political elites who are imposing these restrictions on their businesses and their way of life are not taking it seriously,” he said. “And therefore, these people who often don’t have technical degrees, say ‘COVID isn’t real.’” 

Further complicating matters is the virus’ huge variability in symptoms and outcomes – for instance, while it tends to be deadlier to the elderly, 74-year-old President Donald Trump appeared to bounce right back from it. “For many people, they’re going to heavily weigh the personal experience of the people near them and if they had severe or non-severe cases,” Srinivasan said.  

With the press, politicians and public health officials sending wrong and/or mixed signals, COVID-19 wasn’t just a health crisis, it was a crisis of epistemology: that is, the way society sifts knowledge and makes sense of it

“Our information supply chain has been completely corrupted,” Srinivasan said. Right now people have two choices: Trust the state or trust no one. COVID-19 showed the perils of the former, but the latter is a dead end as well.

“You’re implicitly trusting someone every time you buy some food that it’s not contaminated,” he said. “A truly low trust society is a horrible place to live, because everybody’s trying to cheat.” 

Of course, as someone who aspires to build and not just criticize, he sees a fix. And yes, it involves a blockchain.

I struggled to keep up as Srinivasan laid out his vision for a grand “ledger of record” that would aggregate various data feeds to improve the quality of public information. With the cinematic Iron Man’s laboratory as his Zoom background, he often veered into esoteric (well, to me) tangents about math, physics and computer science. But I think I grasped the gist of it.

To be clear, the system he described is, in his estimation, decades away (2040). Also, his use of the term “on-chain” is expansive, meaning just a decentralized network that offers cryptographic guarantees.

One basic characteristic the ledger of record would have in common with today’s blockchains is that in addition to being open-source (anyone can review the code) it would be open-state (anyone can see the data) and open-execution (anyone can review the instructions carried out by a computer program). 

Further, just like in Bitcoin, anyone could be a “root user,” meaning they would all have the same privileges when accessing the database, unlike a centrally administered one. 

Under such conditions, everyone can agree on what’s in the database, even if they disagree about what it means.

“Whether you’re Indian or Pakistani, Israeli or Palestinian, Japanese or Chinese, Democrat or Republican, everybody agrees on the state of the Bitcoin blockchain,” Srinivasan said.

Oracles, similar to those powering the Chainlink network, would provide data feeds. Their work could be subsidized by betting markets, similar to prediction markets like PredictIt or Augur, except in this scenario their main purpose is not divining the future but verifying the past. Since money is on the line, oracles are incentivized to verify outcomes accurately so bettors, and parties to other types of smart contracts, continue hiring them.

A layer above the oracles would sit “advocates” who would examine, interpret and opine on the data collected. These could include scientists or journalists, but they would make no pretense of objectivity and openly declare their biases. 

See also: Coronavirus: IBM Launching Blockchain ‘Health Pass’ to Support Return to Public Spaces

“When you’re explicit about it, it makes you more trustworthy,” Srinivasan argued. “‘‘I am an advocate for X, discount what I’m saying appropriately. I have skin in the game on Y, W and Z, I have these disclosures, and I’m going to try to convince you of my position with data.’”

That data could include, for example, all the observations and calculations that went into a scientific paper (not just a PDF of the paper itself). Citing a paper would be like importing code. 

“Now, when someone says ‘scientists say,’ you say ‘on-chain or GTFO,’” Srinivasan went on, using an internet acronym meant to express incredulity. Show me the receipts, in other words.

The early version of this digital-age Akashic record already exists, in the form of the on-chain facts we can reason about, he added. That includes every transaction on the Bitcoin and Ethereum networks, everything that can be calculated from them, and every cryptographic fingerprint, or hash, that was recorded on-chain to prove something existed. The ledger of record would take it to the next level.

Had such a system been in place when COVID-19 hit, Srinivasan suggested, it might have been easier to verify, say, the staggering videos circulating in February that purported to show Chinese authorities welding shut Wuhan apartment buildings with residents inside, if the images’ metadata could be checked on-chain. 

Instead of relying on data sets balkanized among hospitals, scientists could have aggregated granular patient information into a global “big table,” which might have helped them figure out why the virus had such varying impacts from place to place and person to person. (This can and should be done in a privacy-preserving way, Srinivasan was quick to add.) The blockchain itself would encourage presenting the data in a standardized way, much as Bitcoin forces competing crypto exchanges to use the same data fields. 

It should be noted that just because a piece of data is recorded on a blockchain, that does not necessarily mean it is true; it simply means the data existed at a particular time in a particular form and was signed by a particular individual (or at least, someone controlling a particular private key). 

But something as seemingly prosaic as a timestamp can save an innocent from going to prison, and recording such metadata on a blockchain makes it difficult to tamper with (“immutable,” in industry parlance).

As he plots his next move after his sabbatical, Srinivasan sounds more optimistic than ever about this technology’s future.

He argued that it is in the same stage as Web 2.0 was in the early 2000s, when few companies were making money off the internet and desktop computing incumbents thought they could safely ignore it.  

“Now what we’ve got is something similar, where crypto is sort of off to the side of tech. It overlaps with it, but it’s still really not taken seriously by lots of people in tech,” Srinivasan said. Yet, the aforementioned advantages of open state and open execution could prove as transformative as open-source code was in that earlier era, he added. 

On a more philosophical note, Srinivasan argued that crypto embodies the core values of technology: it’s international, decentralized, capitalist, volatile, ambitious and “quietly revolutionary, meaning you’re not standing out on the street corner. Revolution comes from a million individual voluntary actions in the privacy of one’s home or one’s phone.”

To top it all off, crypto is a remote-first community, the merits of which became apparent in 2020. 

For all these reasons, Srinivasan predicted, “Crypto is what replaces Silicon Valley.”

Listen to the full audio recording of the CoinDesk interview with Balaji Srinivasan (if you can bear the author’s stammering) and read a rough (AI-generated) transcript here.