Arthur Breitman and Tezos: Why Most Blockchains Are Just Selling Stories

December 2024. While crypto Twitter debates the merits of yet another Layer-1 blockchain promising to solve some niche problem, Arthur Breitman is doing something nobody saw coming: tokenizing physical uranium.

Not uranium futures or derivatives. Physical uranium oxide, the stuff that powers nuclear reactors, available for pennies on Etherlink, Tezos's EVM-compatible Layer-2. The token, XU308, launched on uranium.io and represents actual commodity inventory you can trade in fractions of an ounce.

For Arthur, it makes perfect sense. Because unlike most of crypto, it solves a problem that actually exists.'

The Narrative Problem

Arthur has spent a decade watching blockchains multiply. Most of them, he'll tell you flatly, have no reason to exist.

"Most chains have no reason to exist. Special purpose chains largely exist to sell tokens. In order to sell tokens, you need to have a narrative. And today, if your narrative is going to be, I'm going to be a general purpose chain, it's very hard because people look at you and say, but look, Ethereum is a big general purpose chain. Solana is a big general purpose chain. How will you stand out?"

The problem is that focusing on a specific use case doesn't actually make your technology better. It's marketing dressed up as architecture. A blockchain for supply chain doesn't need its own consensus layer. It needs a token to sell, so it builds a chain to justify it.

He uses an old metaphor: "a toll booth in the middle of an open field." If there's no reason you can't go around it, why would anyone pay the toll?

Even successful projects fall into this trap. Take Uniswap. Most of its actual revenue flows to the company through the front-end, not to token holders. The front-end owns distribution. The contract is commodity. Or consider how chains turn weaknesses into strengths: when Ethereum fees spiked in 2021, the narrative became "Ethereum is a blockchain that people want to pay for."

Arthur also dismisses the utility argument for L1 tokens. If your token needs inflation to pay validators, that's not a use case. That's a cost center.

Arthur started working on Tezos in 2014, after a decade in finance at Goldman Sachs and Morgan Stanley, plus time in robotics and at Google X. But he misjudged something important: scaling.

"I got the importance of scaling wrong. I thought essentially that the main use case was going to be digital gold, a few transfers, and I thought scaling would happen largely through state channels, and it turns out that state channels were not the right way to scale."

State channels didn't work. Rollups did. But his core thesis held: general-purpose infrastructure beats special-purpose chains. You want a system where different use cases collide in unexpected ways.

Governance That Actually Binds

Tezos launched in 2017 with self-amendment at its core. Since then, it's shipped 19 forkless upgrades, each voted on and executed automatically by token holders and validators (called "bakers"). The most recent, Rio, went live in May 2025 with one-day governance cycles and reallocated 10% of staking rewards to the Data Availability Layer, explicitly funding rollup infrastructure.

Most blockchains claim to be decentralized but aren't. Ethereum uses "fork-based governance," where you're theoretically free not to install an upgrade. Except if you don't follow the fork that Circle and Tether support, your chain is dead.

"Ethereum does fork-based governance. If USDC supports that fork, you're done. All DeFi is going to crash. It's governance based on clout. It's extremely centralizing."

Tezos does the opposite. Validators can vote their conscience without worrying about ending up on the wrong fork. The protocol handles upgrades automatically. Very few blockchains do this. Polkadot comes close. Almost nobody else has binding, automated governance.

Arthur co-founded Tezos with his wife, Kathleen, who joined in 2016. The early years were turbulent: the 2017 ICO raised over $230 million but triggered lawsuits settled in 2020 for $25 million, plus governance disputes within the Tezos Foundation. But the protocol kept upgrading.

Etherlink and the Anti-Token Play

In 2023, Tezos launched Etherlink, an EVM-compatible Layer-2 that settles to Tezos. It's fast, cheap, and fully compatible with Ethereum tooling. But it doesn't have its own token. It uses XTZ, Tezos's native token, as gas.

In an industry where every Layer-2 launches with a token to capture value, this is almost unheard of. Etherlink routes activity back to the L1.

"What you don't want is to pay incentives, get mercenary capital, spike your TVL for a short bit, then stop giving incentives and all of that is gone."

In Q2 2025, Etherlink processed roughly 20.5 million transactions, up 474% quarter-over-quarter, representing about 64% of total Tezos transaction volume. The DeFi primitives are there: Uniswap (via Oku), Gearbox, integrations with KyberSwap and Jumper. New products like Midas's mMEV focus on structural yield rather than subsidy farming.

Etherlink is built on Tezos Smart Rollups and is non-custodial with no multisig backdoor. Multiple execution environments are coming: JavaScript, Python, Java, not just Solidity. The Tezos X roadmap aims to condense the million-transaction-per-second capacity demonstrated across parallel rollups in 2021 into a monolithic rollup with unified UX.

Arthur's test for scaling is simple: in a real scalable system, more validators mean higher throughput. Most blockchains don't work that way. Tezos is trying to change that.

Why Uranium Actually Makes Sense

Back to uranium. Why tokenize a commodity most people can't buy?

Because they can't buy it. That's the point.

"The only place where you can globally trade in small amounts physical uranium on earth is uranium.io. You could trade a fraction of an ounce. You could trade it for pennies."

If you want gold, you have options: bullion, ETFs, vaults, at least 20 tokenized projects. Uranium is different. It's strategic. AI data centers need power. Nuclear is expanding. Uranium demand is rising. But spot trading requires million-dollar minimums through specialized brokers.

XU308, organized by Trilitech (Arthur's London-based company), changes that. More metals are coming: hafnium (nuclear control rods), cobalt, lithium, rare earths.

Arthur's thesis is straightforward. If AI makes intelligence cheap, value shifts to physical things: land, energy, minerals.

"If the cost of intelligence goes down, the only stuff that becomes valuable is stuff. So like stuff might be land, energy, minerals, for a while at least human attention."

It's not sexy but real. And in a market drowning in narrative-driven tokens, real is differentiation.

Trilitech is also supporting institutional products: Midas launched tokenized funds on Etherlink (mMEV, mRe7YIELD). Lyzi, a payments company, raised €1.3 million to expand Tezos payment acceptance. Arthur isn't chasing the $10 trillion RWA forecasts some firms throw around, he's more conservative, closer to estimates around $1-4 trillion by 2030.

Beyond Finance

Tezos built a real presence in NFTs through platforms like objkt.com and fxhash. Ubisoft launched Quartz on Tezos. Manchester United had a partnership that ended mid-2025. But Arthur resists being pigeonholed.

"We have a huge art community and it's fantastic. But it doesn't make sense for Tezos to be just an art chain. You don't need an entire chain just for art, it can do a lot more."

Gaming is another focus, but not play-to-earn pyramid schemes. Arthur uses poker as an analogy: money makes the game more fun because it creates stakes. A triple-A game called Reaper Actual is launching on Tezos, betting that blockchain works for in-game economies where ownership matters.

Arthur also wants to drop the "DeFi" label entirely. 

"We should abandon the concept of DeFi and just call it finance. We don't refer to cars as horselescarriages." 

The decentralization will become implicit.

The Long Game

Arthur's vision is clear.

"Scaling without compromise for applications that matter. If you want a thesis for Tezos or a narrative, I think that should be it."

Not scaling by fragmenting value across a dozen L2 tokens. Not scaling by centralizing the sequencer. Scaling by building infrastructure that works, governed transparently, funded sustainably through protocol economics like the DAL allocation.

Tezos is nearly eight years old. It doesn't have Solana's transaction volume or Ethereum's TVL. But it has 19 forkless upgrades, binding governance, an L2 that routes value back to the L1, and a founder willing to say no to the narratives everyone else is chasing.

In late 2024, Tezos launched a prediction market. Polymarket dominated the 2024 election cycle. Tezos is late to the party. But Arthur isn't worried about being first. He's worried about being right.

He's not chasing TVL with mercenary incentives. He's not rebranding weaknesses as strengths. He's building for the moment when we stop saying "blockchain" and just say "infrastructure."

That's the bet. General-purpose infrastructure, governed transparently, scaling without compromise, for applications that actually matter. It's not a narrative. It's just what happens when you spend a decade building instead of selling stories.