It may be that no one can truly understand what Vitalik Buterin has been through in the five years since writing, at age 19, the white paper describing the Ethereum protocol. As he worked with a group of collaborators to flesh out his design, he came to be seen as not only the inventor of Ethereum but something like its high priest, invested with no formal authority yet wielding tremendous soft power. Even critics say he has borne the responsibility with a maturity beyond his years.

The first major attempt to go beyond bitcoin, the Ethereum network now boasts thousands of developers, and its cryptocurrency, ether, has a market capitalization of more than $21 billion, which almost certainly makes Buterin one of the wealthiest people in history to have no discernible ego. The New Yorker called him “indifferently rich,” and indeed it is difficult to imagine a humbler or more unprepossessing centimillionaire. In March, when the price of ether was about $377 (it’s now hovering around $209), Forbes pegged Buterin’s net worth at between $100 million and $200 million.

The success of Ethereum—the network that launched 1,000 coins—has made Buterin not only rich but famous. (William Shatner recently gave him a thumbs-up.) Yet in an age dominated by hard-charging CEOs and the winner-take-all strategies of Uber and Amazon, Buterin seems to be motivated by higher ideals. His peripatetic lifestyle, spending short stints in one country after another, permits little in the way of material possessions. Just as he depends on the kindness of friends for his temporary abodes, so he seems not to have an unkind bone in his own body. His sincerity is touching. Zooko Wilcox, the founder of Zcash, says Buterin once told him, in the early years of bitcoin, “This is the first technology I’ve ever loved that loves me back.”

Yet if the once painfully shy Buterin has played a crucial role in building the sort of community to which he might wish to belong, it is now an ecosystem ambivalent about his continued prominence. I saw the Vitalik Effect firsthand recently at Devcon, the annual Ethereum developer conference that took place in Prague this year. On the first day, a group of attendees spotted Buterin, who was scheduled to give a keynote address the following morning, standing nearby—unprotected, as it were. They began wondering aloud why more people weren’t approaching him, as though Ethereum’s creator were a rock star who should be mobbed by groupies.

“Maybe they’re respecting his space,” one mused.

“No,” said another, “I think it’s because they’re afraid.”

They clearly yearned to go over and introduce themselves, but were at a loss for how they might hold Buterin’s interest. “What would we say to him? We’re application layer [devs]! ‘Thanks for your protocol, we really like it, we think it’s great’?”

They had reached an impasse. But one developer was working up the nerve. “If you dare me,” he said, “I’ll go up and talk to him.”

“I don’t want to dare you,” his friend replied.

In the end, a few of them did troop over and say hi, but only one member of the group had the guts—or the temerity—to shake Buterin’s hand.

That is more than I manage. When I meet Buterin, shortly after his keynote, his hands are full—a mobile phone, a mug of tea—and so he can extend only a single long finger, E.T.-like, by way of greeting. He is tall and almost shockingly lean, and while we talk he sits slightly hunched, often looking down while he ponders a question. When he does lift his gaze to meet mine, his blue eyes radiate warmth; once you have earned his eye contact you want to earn it again.

He toys absently with the string of his tea bag while he speaks. Our brief conversation moves from his efforts to reduce his own importance in the Ethereum ecosystem to the leadership mistake he shares with Donald Trump.

People have called you a philosopher-king—and a while back, when there was a hoax that you’d died, the price of ether plummeted. Are you troubled by the community’s continued reliance on you?
If the community does continue to rely on me, then I think that would definitely be a problem. The whole point of decentralization is that you can make a system where you don’t need to know which specific people are involved in it and that they’re trustworthy in order to be able to participate in it. So if the de facto assumption for Ethereum’s continued existence is that I do certain specific things, then that’s a big risk to anyone in the Ethereum ecosystem—and obviously a large loss of freedom for myself.

At the same time, though, the extent to which I am a critical node in some sense is definitely much lower than it was a year ago. If you look even at some of the Twitter responses from yesterday, people were talking about how it’s kind of clear that the community is relying on each other much more. That’s something that we have been very deliberately working toward for the last year. So even something like, for example, the Ethereum Foundation’s grant program, and [also] how we went with this multi-client approach to implementing Serenity and sharding. One of the goals there was to make it so that the work isn’t just concentrated in a small group of people. You have all these different teams in different places all over the world trying to keep building.

The head of the Python team at the Ethereum Foundation talked today about “onboarding the next one million developers.” Is that the strategy to reduce your prominence—not to shrink your own role but just to bring more and more people on?
Bringing more and more people on is definitely the biggest part of it. If you bring more and more people on, it’s hard to retain the same level of prominence unless you’re in a position where you’re actively managing all of them.

What would be the timeline for that? After Ethereum 2.0, Serenity, is implemented?
It’s just something that’s going to naturally happen over time.

The slide you showed of the Hindenburg disaster to illustrate the DAO hack was funny. And it made me think about how you led the effort to slow down the attacker and roll back the effects of his actions. Aren’t there some advantages to being what Satoshi once was for bitcoin and what Linus Torvalds has been for Linux—”Benevolent Dictator for Life”?
There definitely are. It is a good thing for a development community to be smaller and more concentrated especially in the earlier stages, where there are lots of decisions to make. Designing big protocols by large committee is something that just doesn’t work. But when the system stabilizes and we’re talking less about huge fundamental revamps and more about ongoing marginal tweaks, then I think that’s the point where the [decentralized] model starts both working more and making more sense as the right way to do it.

One of the earliest ideas with blockchain was that it means you don’t have to trust people; you can just trust math. Maybe that was true for bitcoin early on, but when it comes to things like super-quadratic sharding, cross-shard transactions, SNARKs, and so on—for those who can’t read code or verify the validity of these concepts, are they not forced once more to accept the decisions and judgements of a sort of priestly class?
There’s definitely a balance there. Even in bitcoin, for example, [the cryptographic hash function] SHA-256—you might think of it as one thing, because it’s one word and you never peek into the blocks. But really it’s an insanely complicated thing with a pedigree of decades of academic research that just completely destroyed the security of a whole bunch of alternatives. On the other hand, this is part of the reason why I’m not a big fan of super-quadratic sharding at this point. We’re just doing quadratic sharding. I do feel that [implementing the proof-of-stake system] Casper is necessary to give Ethereum its higher level of security and efficiency; sharding is necessary to bring it up to the level of scalability that we need. But you do have to be very hawkish about protocol simplicity. And there are areas where I’m trying hard to come up with ways to push the protocol’s complexity down.

One example is that there’s this Merkle-tree structure in the current Ethereum chain, and it’s this fairly complicated thing—basically it’s a way of storing the data about all of the accounts in the system. I’ve been thinking a lot about how to cut the complexity of that by maybe a factor of five. But we’ll see. Reducing the number of lines of code that you need to worry about is important.

Are there any other considerations?
There’s also conceptual simplicity. But even there you have to think a bit more carefully. Because, for example, the bitcoin protocol seems simple, but the analysis in many ways is not simple. For example, people thought [from 2009 to 2013] that the bitcoin security [threshold] was 50 percent. But then Emin Gün Sirer and Ittay Eyal wrote the “selfish mining” paper and, oh, my God, now it’s between zero and 33 percent, depending on your assumptions about the network, and here’s a patch that the bitcoin core devs never implemented that will make it 25 percent. [Editor’s note: In their 2013 paper, Sirer and Eyal proposed a fix to the bitcoin protocol that was intended to raise the network’s security threshold—the amount of total computational power in the network one would have to control in order to double-spend bitcoins or obtain more than one’s fair share of mining rewards—to 25 percent.] So now you have to [consider] your attitudes about the possibility of collusion in the mining network, and your opinions about how the peer-to-peer network works, and a bunch of other things. Whereas [Ethereum’s] proof-of-stake algorithm, we’ve definitely put more work into it. And in some respects the analysis [of its security] will be simpler. There’s also just a much larger community that’s doing the analysis.

So it’s a little easier to trust because it’s not just five people telling you to take it on faith or something?
Yeah, yeah. I definitely keep an eye out for solutions that seem like [they require] taking five people on faith, and I try hard to avoid them.

The fifth anniversary of your Ethereum white paper is coming up in November. Your father recently said that you were “innocent and unprepared” when you co-founded Ethereum and that you “had to learn a lot of tough lessons about people.” What’s the hardest lesson you’ve had to learn?
One is that it’s easy to overestimate the extent to which everyone thinks the way that you do. Back in 2014 I thought, at least, that I was in the space because I believed in decentralization, believed in making open, public things for the world, believed in censorship-resistant mutual platforms, and all of that. There were a bunch of debates which had to do with the fact that the other people on the [Ethereum] team just basically wanted to make a huge pile of money. Not all of them, but some. At some point I realized that you can’t resolve all conflicts by just getting the two people to sit down and have a conversation with each other and hug each other. That resolves it for one hour, but then they go back to their rooms, and if the underlying issues aren’t addressed then it’s not changed.

You learned that dialogue has a place, but it can’t solve everything?
Yeah, exactly. The other big one is that a lot of people were nice to me, and I thought they were nice to me because they were nice, but really they were nice to me because they perceived I was powerful. One of the mistakes I made as a leader at that time is, actually, the same mistake I noticed Donald Trump making a while ago. If you remember the time when he was really in favor of blowing up Obamacare, and then Barack came over the White House, they talked for one and a half hours, and then right after that, Trump said, “Oh, my God, I never believed healthcare could be so complicated.” But then, of course, after that he wasn’t particularly [in favor of] trying to stop blowing up Obamacare. The trait that I think about here is: agreeing with the last person you talked to. It’s actually very easy to do that if you don’t have experience. It did take me about a year or so to figure out how to move past that.

This interview has been edited for length and clarity.

Main image Trevor Jones.