Rift wants to make swapping bitcoin crosschain not suck
The team is betting that a 20-minute hardware trust window beats a new alt-L1
Shutterstock AI/Shutterstock and Adobe modified by Blockworks
This is a segment from The Breakdown newsletter. To read full editions, subscribe.
Most cross-chain protocols want you to trust a validator set. Rift wants you to trust a chip — and only for 20 minutes.
The startup is launching a peer-to-peer bitcoin-to-EVM trading protocol that runs full nodes for Bitcoin and Ethereum inside a trusted execution environment (TEE). The idea is simple: If the enclave can see that both legs of the trade landed, it lets go of the funds. If not, it doesn’t.
“We use [TEEs] to match orders as well as validate that someone got paid,” CEO Samee Siddiqui told me. The TEE acts as an escrow for the funds until sufficient block confirmations indicate it’s safe to release.
That shifts the core trust model, compared to what’s available today. There are no multisigs or distinct proof-of-stake chains involved, or the need to mint a synthetic BTC somewhere. Just native bitcoin, held in a hardware enclave, for a time-boxed window while the other leg of the trade settles.
Yes, that window is a risk. “If the machine blows up, then anything in those 20 minutes is potentially lost,” Siddiqui admitted. But that’s the design tradeoff: minimal trust and minimal attack surface for a minimal duration.
Why do this? To make swaps faster, cheaper and less reliant on staked capital.
“Most people don’t honestly care that much about security,” Siddiqui said. (That’s a pity, but clearly true.) “The pitch is actually [that] this is the most capital-efficient solution,” he said.
The fee math backs that up: Rift charges 10 bps taker, 0 maker fees. That’s a big drop from the 40-60 bps you’ll find on Coinbase, of course. But it’s also cheaper than THORChain’s RUNE-bonded AMM model and avoids the economic security game theory with no governance token overhead.
Instead, market makers provide the liquidity and handle their own rebalancing — often using cbBTC as the ETH-side routing asset. That’s for “liquidity efficiency” at the get-go, Siddiqui said, but noted that USDT and other pairs are coming.
It’s not the first time someone’s tried to make BTC↔ETH swaps trustless. RenBTC had its moment (until it didn’t). Threshold Network is still going strong. Aside from THORChain, there’s also Near Intents as an option.
Rift tried other paths first, including a zero-knowledge BTC light client. “But we ended up going with [TEEs] because it’s a lot more flexible,” Siddiqui said. He also pointed out zk stack complexity: “Auditing that a zero-knowledge proof is correct is actually much more difficult than auditing a TEE is correct.”
The go-to market isn’t consumer-first — it’s API-first. Rift wants to plug into wallets and swap interfaces that already have volume. The more flows they can sit behind, the stickier they get. Think DEX aggregators, not a dedicated frontend.
The bet is that onchain bitcoin trading doesn’t need a new L1 or a wrapped token. It just needs a swap protocol with the right capital efficiency, UX and trust profile.
In Siddiqui’s mind: “If we don’t have DEXs flip CEXs, then the entire space, basically, is a failure.”
Get the news in your inbox. Explore Blockworks newsletters:
- The Breakdown: Decoding crypto and the markets. Daily.
- 0xResearch: Alpha in your inbox. Think like an analyst.
