Best of x.com/drakefjustin on Kiwi News
@drakefjustin on X · Most upvoted by Kiwi News readers
16 stories by @drakefjustin on X have been shared on Kiwi News since 19 May 2024, with 43 upvotes and 7 comments in total. The latest was shared on 7 October 2026. They are listed below, most upvoted first.
Text under a title is an AI summary of the linked article.
- Justin Drake: „I dropped the EigenLayer advisorship“
Justin Drake says he dropped his EigenFoundation advisorship in September and apologized to the Ethereum community and EF colleagues for the drama he caused, calling it a bad move in hindsight. He says he will now turn down all advisorships, angel investments, and security councils, a personal policy that goes beyond the recent EF-wide conflict of interest policy to signal commitment to neutrality. Drake also stepped out of the ultra sound team to simplify and refocus, saying Alex, Niclas, and Christian have taken complete responsibility for its operations and decision making.
- ETH is 10 years old today!
The article marks the tenth anniversary of Ether (ETH), whose ICO began July 22, 2014, when ETH was sold at 2,000 ETH per 1 BTC without VCs or vesting. It states that 1 BTC now buys less than 20 ETH, and that few assets have outperformed BTC over 10 years, with fewer doing so by 100x. The author cites 4.3M ETH burned for gas since EIP-1559, staking that provides $100B of economic security (10x Bitcoin's), and US spot ETH ETFs starting to trade the following day.
- @drakefjustin: Google and Oratomic papers improve Shor's algorithm for secp256k1
Google Quantum AI and Oratomic have each published papers improving Shor's algorithm against secp256k1, the elliptic curve used for Bitcoin and Ethereum signatures, and the author says the results compound. The Google paper estimates the logical circuit runs on about 1,200 logical qubits, roughly 500K physical qubits on a superconducting computer with surface code error correction, and could recover a key in minutes. The Oratomic paper, which the author describes as a stealthy startup with ex-Google and Caltech faculty, estimates 26,000 atomic qubits suffice, about 40x fewer physical qubits than the previous state of the art, but a single run would take about 10 days on neutral atom hardware.
- My bat signal 🦇🔊 will return when ETH is ultra sound again, soon enough™
The author argues that ETH will become ultra sound again, with supply shrinking, because issuance will fall and burn will rise, and that Bitcoin's proof-of-work security is untenable long term. On ETH issuance, the author says the current curve guarantees a 2% tail APR even at full staking and proposes "croissant issuance," a half-oval with a 50% staking soft cap and a peak of about 1%/year. On burn, the author expects blob data availability demand to grow, with blob burn possibly reaching hundreds of ETH per day this year before falling again after the Fusaka fork.
- @drakefjustin: blockchain industry should plan for ECDSA break, migrate to fresh
Justin Drake urges the blockchain industry to calmly plan for "bunker mode," chiefly by migrating assets to fresh addresses whose public keys stay hidden behind a hash, and he says ECDSA could break within months rather than years. He argues that recent advances in AI-driven mathematics, including OpenAI's mathematical results, make such a break plausible, and he defines "break" as fast private key recovery, for example in one week on a large GPU cluster. He cautions against a rushed migration and says Wallets holding under 50 BTC enjoy partial cover from "Satoshi's shield," the roughly 20K exposed addresses holding 50 BTC each. He also recommends hash-based cryptography as the path after bunker mode, with the Ethereum roadmap on strawmap.org to be accelerated accordingly.
- @drakefjustin: Ethereum Foundation abandoning Poseidon for SHA/BLAKE on L1
The Ethereum Foundation is abandoning Poseidon for L1 in favor of SHA2 or BLAKE, according to a post by @drakefjustin, because breakthrough SNARK designs show traditional hash functions can now match Poseidon inside a SNARK. The post credits binary-field research, including Binius by Jim and Ben in 2023 and Flock by Ron, Benedikt and William in June, for making this possible. It claims binary-field SNARKs reach 1M traditional hash calls proven per second on a laptop, with about 100x overhead versus native CPU boolean compute. The strawmap points to a production-grade leanVM in 2027 and CL, DL, EL deployments in 2028.
- @drakefjustin: Ethereum Foundation forms Post Quantum team lead Thomas Coratger
The Ethereum Foundation has formed a new Post Quantum (PQ) team led by Thomas Coratger, with Emile, a contributor to leanVM, joining him, and says EF management has declared PQ security a top strategic priority. The announcement lists several initiatives: a $1M Poseidon Prize to harden the Poseidon hash function and a separate $1M Proximity Prize, plus bi-weekly PQ transactions calls led by Antonio Sanso starting next month. Multi-client PQ consensus devnets are now live, coordinated by Will Corcoran via weekly interop calls, and the roadmap for a full transition is due soon on pq.ethereum.org.
- @drakefjustin: Ethereum validator gas limit boost to 60M for Pectra upgrade
- @drakefjustin: Bitcoin PoW security is a ticking time bomb, fees at 13-year low
Justin Drake argues that Bitcoin's proof-of-work security model is unsustainable because fee revenue is falling relative to issuance, leaving the network open to a cheap 51% attack. He cites fees at a 13-year low of under 10 BTC/day and miner fee revenue at about 1% of the security budget, a share he says has held since the 2016 halving. Drake calculates that a permanent 51% attack on a $200T Bitcoin would cost about $20B, or 0.01% of market cap, which he says is within reach given $43B of BTC perpetual open interest. He rejects BitVM, covenants and proof-of-authority as fixes and says tail issuance or proof-of-stake are the remaining options.
- Real-time Ethereum block proving breakthrough
Succinct reports that it now proves mainnet Ethereum L1 blocks in under one 12-second slot, with 94% proven in under 12 seconds, 99% in under 13 seconds, and 99.9% in under 12 seconds, a figure described as coming soon. RISC Zero unveiled a $120K home GPU cluster expected to produce proofs in 9.25 seconds, and Brevis, OpenVM, Snarkify, ZisK and ZKM are said to be weeks from joining them. Regular gas limit increases are also advocated, with 10% of stake reportedly voting for a 60M limit, and Ethproofs call #2 is scheduled for May 23 at 2pm UTC.
- @drakefjustin: why we're building internet finance
Justin Drake argues that internet finance should be built on a set of contrasts: free trade over trade wars, permissionless innovation over bureaucracy, and sovereign individuals over intermediaries. The post frames these as a multi-decade vision rather than a four-year cycle, and it contrasts undebasable money with money printers, cryptography with aircraft carriers, and fairness by design with cronyism. It closes by stating that internet finance is coming and urging readers to believe in it.
- @drakefjustin: Bitcoin's security model is a ticking time bomb
Justin Drake argues that Bitcoin's proof-of-work security model is a ticking time bomb because fee revenue is too small to sustain mining. He cites fees at a 13-year low of less than 10 BTC/day and miner revenue from fees at 1%, with fees falling as block subsidies halved in 2016, 2020, 2022 and 2024. He estimates that at a $1M price, a permanent 51% attack would cost about $20B against a $200T asset. He concludes that proof-of-work is not sustainable without a fix, and that tail issuance or proof-of-stake would be the main candidate solutions.
- @drakefjustin: real-time Ethereum block proving breakthrough
Justin Drake argues that real-time zk proving of every Ethereum mainnet block is now achievable, so Ethereum L1 can become a "based and native rollup" that is also a zk rollup. He says zkVMs are compounding about 10x per year, and that a handful will reach real-time proving by the end of Q2, with diversity rather than formal verification as the near-term security strategy. He also suggests that gas limits could track zkVM gains, reaching 1 gigagas/sec in about three years if 10x annual progress continues, while noting that EIP-9698 proposes roughly 3x-per-year gas bumps.
- The beacon chain genesis happened four years ago on Dec 1, 2020.
Ethereum can achieve a broad set of protocol upgrades and become the settlement layer for the internet of value, according to the text. The beacon chain began on Dec 1, 2020, with 0.5M ETH staked and no immediate user benefits, and the text cites 10K consensus participants, $125B in economic security and 100% uptime. Proposed improvements include better censorship resistance, MEV handling, smaller staking deposits, full danksharding and native rollups, while L2s are expected to deliver performance gains in months rather than years.
- how a from-scratch redesign of the Ethereum consensus layer could look like
A from-scratch redesign of the Ethereum consensus layer will be unveiled tomorrow at 5pm on the Devcon main stage, after a year of development. Its stated goal is a credible strategy to ship an ambitious beacon chain roadmap within a reasonable timeframe. The proposal was developed over several months with researchers and consensus developers, and its proposer says community participation is key if the approach is to reach rough consensus. The session was listed on the Devcon schedule under a redacted title.
- Justin Drake's comment on advisorship in EigenFoundation
Ethereum researcher Justin Drake disclosed that he recently became an advisor to the EigenFoundation, which comes with an EIGEN token incentive he says could be worth more than all his other assets combined, vesting over three years. He pledges to reinject advisorship proceeds into Ethereum projects and to end the advisorship if EigenLayer moves against Ethereum's interests. Drake says his mandate is limited to researching restaking risks, that he will not appear in marketing material, and that he advises in an individual capacity. He also states he will likely keep a critical public stance toward EigenLayer, and that discussions with Sreeram about the role took over a year.