QMS Network Official
The post-quantum transition cannot wait for a perfect signature algorithm. That is the central message of Cloudflare's latest analysis. ML-DSA-44 is available today, but it carries real costs. Its signatures are 2,420 bytes, compared with 64 bytes for Ed25519…
Valuable and timely analysis from the Cloudflare team:
https://blog.cloudflare.com/ml-dsa-will-have-to-do/
https://blog.cloudflare.com/ml-dsa-will-have-to-do/
Cloudflare Blog
Why we cannot wait for better post-quantum signature algorithms
NIST is advancing nine new post-quantum signature algorithms as potential candidates for future standardization. We take a closer look at all of them, and argue that while they are in the works and show great potential, we should use ML-DSA for now — the…
❤2
Meet Egeyar Özlen Saiu, our Lead Protocol Engineer.
Fun fact: the world's most widely used compiler and linker include code she wrote.
Egeyar has spent more than 9yrs building the lower layers of distributed systems, with a focus on consensus protocols and applied cryptography.
Her path to QMS:
→ Core consensus engineering at DFINITY during the design and launch of the Internet Computer
→ Led the development of a distributed secret-management and disaster-recovery system at Lit Protocol
→ Now implementing and hardening QMS’s novel consensus and finality layers
Egeyar joined us in March, and we're truly happy to have her on the team!
Original Tweet
X | LinkedIn | Medium | Docs
Fun fact: the world's most widely used compiler and linker include code she wrote.
Egeyar has spent more than 9yrs building the lower layers of distributed systems, with a focus on consensus protocols and applied cryptography.
Her path to QMS:
→ Core consensus engineering at DFINITY during the design and launch of the Internet Computer
→ Led the development of a distributed secret-management and disaster-recovery system at Lit Protocol
→ Now implementing and hardening QMS’s novel consensus and finality layers
Egeyar joined us in March, and we're truly happy to have her on the team!
Original Tweet
X | LinkedIn | Medium | Docs
❤4
QMS Network Official
Meet Egeyar Özlen Saiu, our Lead Protocol Engineer. Fun fact: the world's most widely used compiler and linker include code she wrote. Egeyar has spent more than 9yrs building the lower layers of distributed systems, with a focus on consensus protocols and…
If you'd like to learn more, here’s Egeyar's LinkedIn profile: https://www.linkedin.com/in/egeyar/
❤2🔥2
Qubits needed to break RSA-2048:
2019: ~20M
2025: under 1M
2026: research claims under 100k
What happens by 2027?
Original Tweet
X | LinkedIn | Medium | Docs
2019: ~20M
2025: under 1M
2026: research claims under 100k
What happens by 2027?
Original Tweet
X | LinkedIn | Medium | Docs
👀3❤2
Think of a quantum computer as a very sensitive musical instrument that constantly slips out of tune.
Until now, engineers often had to stop the computation, retune the machine, and restart it.
Google’s latest research, published July 22, introduces an AI “technician” that listens to a machine’s error signals and adjusts it while it keeps operating.
The breakthrough: the system can detect its own hardware drift and continuously recalibrate itself during computation. This matters because useful quantum computations will eventually need to run uninterrupted for long periods.
Google demonstrated this on Willow in quantum-memory experiments. While not yet with a long, commercially useful algorithm, they’re making quantum hardware more capable of operating like reliable infrastructure.
Original Tweet
X | LinkedIn | Medium | Docs
Until now, engineers often had to stop the computation, retune the machine, and restart it.
Google’s latest research, published July 22, introduces an AI “technician” that listens to a machine’s error signals and adjusts it while it keeps operating.
The breakthrough: the system can detect its own hardware drift and continuously recalibrate itself during computation. This matters because useful quantum computations will eventually need to run uninterrupted for long periods.
Google demonstrated this on Willow in quantum-memory experiments. While not yet with a long, commercially useful algorithm, they’re making quantum hardware more capable of operating like reliable infrastructure.
Original Tweet
X | LinkedIn | Medium | Docs
❤5
"Quantum doesn’t matter yet. The machines aren’t here."
That misses how the threat works.
For encrypted data, the risk is Harvest Now, Decrypt Later: capture ciphertext today, decrypt it when the hardware arrives.
On blockchains, transaction histories, signatures, and exposed public keys remain in a permanent public archive. The harvesting is already done on the ledger.
Replacing vulnerable software with quantum-resistant versions is complicated and takes time. That's another reason we need to act before quantum computers are here.
Original Tweet
X | LinkedIn | Medium | Docs
That misses how the threat works.
For encrypted data, the risk is Harvest Now, Decrypt Later: capture ciphertext today, decrypt it when the hardware arrives.
On blockchains, transaction histories, signatures, and exposed public keys remain in a permanent public archive. The harvesting is already done on the ledger.
Replacing vulnerable software with quantum-resistant versions is complicated and takes time. That's another reason we need to act before quantum computers are here.
Original Tweet
X | LinkedIn | Medium | Docs
❤2👍2
This media is not supported in your browser
VIEW IN TELEGRAM
1️⃣ BlackRock, Fidelity, and Coinbase are now paying for Bitcoin's quantum migration.
On 23 July, nine firms launched the Bitcoin Security Consortium: BlackRock, Fidelity Digital Assets, Coinbase, Strategy, Block, Blockstream, Anchorage Digital, ARK Invest, and Galaxy.
$15M over three years for Bitcoin security researchers and open-source developers. Post-quantum preparation is a named focus. Each member funds independently, and the group takes no position on protocol changes.
Migration on a live chain is a collective-action problem. It moves when the biggest holders start writing checks. Even then, the money funds research, nobody can buy a protocol change.
2️⃣ The same day, Coinbase published its post-quantum custody plan: a PQ successor to the key management system securing roughly 99.9% of the assets it custodies, plus a quantum-safe signing pipeline (secure enclaves + threshold cryptography) due within a year.
Custody can re-engineer itself in a year. The settlement layer underneath needs consensus and years. That asymmetry is the retrofit problem in one picture.
QMS is built around this assumption: the next generation of blockchain infrastructure should be quantum-resistant by design and capable of turning network compute into useful work.
Original Tweet
X | LinkedIn | Medium | Docs
On 23 July, nine firms launched the Bitcoin Security Consortium: BlackRock, Fidelity Digital Assets, Coinbase, Strategy, Block, Blockstream, Anchorage Digital, ARK Invest, and Galaxy.
$15M over three years for Bitcoin security researchers and open-source developers. Post-quantum preparation is a named focus. Each member funds independently, and the group takes no position on protocol changes.
Migration on a live chain is a collective-action problem. It moves when the biggest holders start writing checks. Even then, the money funds research, nobody can buy a protocol change.
2️⃣ The same day, Coinbase published its post-quantum custody plan: a PQ successor to the key management system securing roughly 99.9% of the assets it custodies, plus a quantum-safe signing pipeline (secure enclaves + threshold cryptography) due within a year.
Custody can re-engineer itself in a year. The settlement layer underneath needs consensus and years. That asymmetry is the retrofit problem in one picture.
QMS is built around this assumption: the next generation of blockchain infrastructure should be quantum-resistant by design and capable of turning network compute into useful work.
Original Tweet
X | LinkedIn | Medium | Docs
🔥3❤2
QMS Network Official
1️⃣ BlackRock, Fidelity, and Coinbase are now paying for Bitcoin's quantum migration. On 23 July, nine firms launched the Bitcoin Security Consortium: BlackRock, Fidelity Digital Assets, Coinbase, Strategy, Block, Blockstream, Anchorage Digital, ARK Invest…
Keeping an eye on SBC'26 at Stanford this week: Monday's quantum threat session (Google, Ledger, Circle), Tuesday's talk on the economics of proof-of-useful-work, and Friday's affiliated Bitcoin cryptography workshop on post-quantum signatures.
👀3❤2
Even if quantum computers never break a single Bitcoin key, preparing for the threat still comes with two costs.
First, migration.
A Chaincode Labs report estimates that moving the full UTXO set into quantum-resistant outputs could take 305 to 568 days if migration receives 25% of block space.
Then, ongoing overhead.
Depending on the scheme Bitcoin adopts, post-quantum signatures can be roughly 10x to more than 100x larger than today’s 64-byte Schnorr signatures.
The exact bill depends on the design. But retrofitting a live monetary network consumes block space, fees, bandwidth, and coordination before the first key is ever broken.
Original Tweet
X | LinkedIn | Medium | Docs
First, migration.
A Chaincode Labs report estimates that moving the full UTXO set into quantum-resistant outputs could take 305 to 568 days if migration receives 25% of block space.
Then, ongoing overhead.
Depending on the scheme Bitcoin adopts, post-quantum signatures can be roughly 10x to more than 100x larger than today’s 64-byte Schnorr signatures.
The exact bill depends on the design. But retrofitting a live monetary network consumes block space, fees, bandwidth, and coordination before the first key is ever broken.
Original Tweet
X | LinkedIn | Medium | Docs
❤3
How Proof-of-Useful-Work Rewards the QMS Community
Proof of work has always had one economic problem: the work itself has no value beyond securing the chain. Miners spend energy on hashes no one else wants. The protocol pays for them because no one else will.
QMS keeps the security logic of PoW and puts a second customer beside the protocol. Miners run solvers on real optimization problems posted by enterprise clients, and the same solver run earns from both. One computation, two revenue streams.
The loop is bigger than miners. Finalizers stake the native token and earn a share of every minted block for making history final. Clients pay only for solutions that clear their own bar. Users transact on security that useful work helps fund. Holders sit behind three burn flows — base fees, a fraction of every solution reward, and every unearned finality share — designed to offset dilution as the marketplace grows.
This article covers:
→ The three income streams a QMS miner earns — block rewards, EIP-1559 priority fees, and client solution payments
→ How finalizers, clients, users, and holders each hold a position in the same loop
→ The Ofelimos-based construction that ties every payout to verifiable work — seeded search, checkpoint trails, and a quality-blind lottery
→ Why QUBO is the launch problem class, and how quantum annealers and QAOA machines plug into the same math natively
Every token paid on QMS traces back to work that really happened. When quantum hardware clears client thresholds, it earns through the same marketplace — no protocol change required.
Read the full piece here: https://qms.finance/news/how-proof-of-useful-work-rewards-the-qms-community
X | LinkedIn | Medium | Docs
Proof of work has always had one economic problem: the work itself has no value beyond securing the chain. Miners spend energy on hashes no one else wants. The protocol pays for them because no one else will.
QMS keeps the security logic of PoW and puts a second customer beside the protocol. Miners run solvers on real optimization problems posted by enterprise clients, and the same solver run earns from both. One computation, two revenue streams.
The loop is bigger than miners. Finalizers stake the native token and earn a share of every minted block for making history final. Clients pay only for solutions that clear their own bar. Users transact on security that useful work helps fund. Holders sit behind three burn flows — base fees, a fraction of every solution reward, and every unearned finality share — designed to offset dilution as the marketplace grows.
This article covers:
→ The three income streams a QMS miner earns — block rewards, EIP-1559 priority fees, and client solution payments
→ How finalizers, clients, users, and holders each hold a position in the same loop
→ The Ofelimos-based construction that ties every payout to verifiable work — seeded search, checkpoint trails, and a quality-blind lottery
→ Why QUBO is the launch problem class, and how quantum annealers and QAOA machines plug into the same math natively
Every token paid on QMS traces back to work that really happened. When quantum hardware clears client thresholds, it earns through the same marketplace — no protocol change required.
Read the full piece here: https://qms.finance/news/how-proof-of-useful-work-rewards-the-qms-community
X | LinkedIn | Medium | Docs
❤3
Protocol-wide post-quantum signatures deployed by major L1s:
0
Bitcoin has draft BIPs, Ethereum and Solana have roadmaps. The progress is real but production migration is still pending.
As Q-Day approaches, the gap between "planned" and "deployed" becomes the risk.
Original Tweet
X | LinkedIn | Medium | Docs
0
Bitcoin has draft BIPs, Ethereum and Solana have roadmaps. The progress is real but production migration is still pending.
As Q-Day approaches, the gap between "planned" and "deployed" becomes the risk.
Original Tweet
X | LinkedIn | Medium | Docs
❤3
A chain where only miners earn doesn't grow. QMS gives every participant a stake in the loop.
Miners and finalizers earn directly. Clients, users, and holders gain from the same loop without running hardware.
Original Tweet
X | LinkedIn | Medium | Docs
Miners and finalizers earn directly. Clients, users, and holders gain from the same loop without running hardware.
Original Tweet
X | LinkedIn | Medium | Docs
❤2
QMS Network Official
How Proof-of-Useful-Work Rewards the QMS Community Proof of work has always had one economic problem: the work itself has no value beyond securing the chain. Miners spend energy on hashes no one else wants. The protocol pays for them because no one else will.…
Find out more in our latest research article 👆
❤1
Meet the people building QMS.
Allan Brøndum Rasmussen is our Lead Cryptography Engineer.
Allan holds a Master's degree in Mathematics from Aarhus University, covering advanced analysis, number theory, and algebra, followed by a Bachelor's degree in Computer Science. He specialises in high-assurance Rust, cryptographic protocol implementation, and the design of complex distributed systems.
He spent seven years at Systematic A/S as technical lead on large-scale healthcare software, developing strong skills in performance-critical backend systems, database design, and reliable distributed architecture. He then joined Concordium as a Senior Systems Engineer, where he led the rewrite of the blockchain node's execution layer from Haskell to Rust and collaborated with researchers on implementing cryptographic protocols, including zero-knowledge-based identity systems.
Allan joined QMS Network in April, and he's is responsible for ensuring that the security properties established at the research and design level translate faithfully into the implementation.
Original Tweet
X | LinkedIn | Medium | Docs
Allan Brøndum Rasmussen is our Lead Cryptography Engineer.
Allan holds a Master's degree in Mathematics from Aarhus University, covering advanced analysis, number theory, and algebra, followed by a Bachelor's degree in Computer Science. He specialises in high-assurance Rust, cryptographic protocol implementation, and the design of complex distributed systems.
He spent seven years at Systematic A/S as technical lead on large-scale healthcare software, developing strong skills in performance-critical backend systems, database design, and reliable distributed architecture. He then joined Concordium as a Senior Systems Engineer, where he led the rewrite of the blockchain node's execution layer from Haskell to Rust and collaborated with researchers on implementing cryptographic protocols, including zero-knowledge-based identity systems.
Allan joined QMS Network in April, and he's is responsible for ensuring that the security properties established at the research and design level translate faithfully into the implementation.
Original Tweet
X | LinkedIn | Medium | Docs
❤7
QMS Network Official
Meet the people building QMS. Allan Brøndum Rasmussen is our Lead Cryptography Engineer. Allan holds a Master's degree in Mathematics from Aarhus University, covering advanced analysis, number theory, and algebra, followed by a Bachelor's degree in Computer…
Allan’s fun fact lands a little too well: “Just here to get some useful work done.” (pun intended) 😄
❤5😁2
Compute mining’s economic model can carry into a quantum future, even as the machines change.
Quantum processors will enter as specialized accelerators in hybrid classical-quantum systems, rather than as drop-in replacements for today’s mining rigs.
What carries forward is the coordination layer: route useful optimization work to miners and reward qualifying solutions, regardless of the hardware behind them.
In QMS, quantum hardware can first improve a classical miner’s search. Direct quantum mining follows once probabilistic quantum work can be verified securely.
Original Tweet
X | LinkedIn | Medium | Docs
Quantum processors will enter as specialized accelerators in hybrid classical-quantum systems, rather than as drop-in replacements for today’s mining rigs.
What carries forward is the coordination layer: route useful optimization work to miners and reward qualifying solutions, regardless of the hardware behind them.
In QMS, quantum hardware can first improve a classical miner’s search. Direct quantum mining follows once probabilistic quantum work can be verified securely.
Original Tweet
X | LinkedIn | Medium | Docs
❤3
This media is not supported in your browser
VIEW IN TELEGRAM
Post-quantum security moved into the engineering layer this week.
3 signals stood out:
1️⃣ AI found a serious weakness in HAWK under 60 hours.
On 28 July, Anthropic said researchers working with Claude Mythos Preview found a new attack against HAWK, a third-round candidate in NIST’s additional-signature process. The attack cut HAWK’s effective key strength roughly in half, and its authors withdrew the scheme. NIST said its finalized standards, including ML-KEM, ML-DSA, and SLH-DSA, are unaffected.
The standards process worked: HAWK failed before anyone deployed it. The new part is the speed. The scheme had already passed two rounds of expert review before the AI-assisted team found the weakness.
2️⃣ Quantumglow rebuilds parts of consensus around post-quantum constraints.
Anza published Quantumglow on 30 July as a post-quantum version of Solana’s Alpenglow design. Larger signatures and the lack of practical BLS-style aggregation made a direct algorithm swap unworkable. Anza’s answer combines a tailored hash-based signature scheme, block commitments, authenticated channels, local certificate events, and a new approval mechanism.
3️⃣ Cloudflare now supports post-quantum authentication to origin servers.
On 29 July, Cloudflare added ML-DSA support to Authenticated Origin Pulls and Custom Origin Trust Store. Customers can now configure FIPS 204 certificates to authenticate connections between Cloudflare and their origin servers.
Cloudflare had already brought post-quantum key exchange to these connections. ML-DSA now covers authentication too, protecting against future attackers forging classical credentials.
Original Tweet
X | LinkedIn | Medium | Docs
3 signals stood out:
1️⃣ AI found a serious weakness in HAWK under 60 hours.
On 28 July, Anthropic said researchers working with Claude Mythos Preview found a new attack against HAWK, a third-round candidate in NIST’s additional-signature process. The attack cut HAWK’s effective key strength roughly in half, and its authors withdrew the scheme. NIST said its finalized standards, including ML-KEM, ML-DSA, and SLH-DSA, are unaffected.
The standards process worked: HAWK failed before anyone deployed it. The new part is the speed. The scheme had already passed two rounds of expert review before the AI-assisted team found the weakness.
2️⃣ Quantumglow rebuilds parts of consensus around post-quantum constraints.
Anza published Quantumglow on 30 July as a post-quantum version of Solana’s Alpenglow design. Larger signatures and the lack of practical BLS-style aggregation made a direct algorithm swap unworkable. Anza’s answer combines a tailored hash-based signature scheme, block commitments, authenticated channels, local certificate events, and a new approval mechanism.
3️⃣ Cloudflare now supports post-quantum authentication to origin servers.
On 29 July, Cloudflare added ML-DSA support to Authenticated Origin Pulls and Custom Origin Trust Store. Customers can now configure FIPS 204 certificates to authenticate connections between Cloudflare and their origin servers.
Cloudflare had already brought post-quantum key exchange to these connections. ML-DSA now covers authentication too, protecting against future attackers forging classical credentials.
Original Tweet
X | LinkedIn | Medium | Docs
❤2🔥2
QMS Network Official
Post-quantum security moved into the engineering layer this week. 3 signals stood out: 1️⃣ AI found a serious weakness in HAWK under 60 hours. On 28 July, Anthropic said researchers working with Claude Mythos Preview found a new attack against HAWK, a…
The QMS takeaway:
QMS starts from a simple assumption. The next generation of blockchain infrastructure should be quantum-resistant by design, and network compute should do useful work.
This week showed how much harder those choices become once the infrastructure is already live.
QMS starts from a simple assumption. The next generation of blockchain infrastructure should be quantum-resistant by design, and network compute should do useful work.
This week showed how much harder those choices become once the infrastructure is already live.
❤3
AI is turning compute markets into some of the most important infrastructure of the next decade, expanding blockchain’s role in coordinating them.
The demand curve keeps steepening. More models, more inference, more work that has to run somewhere. Microsoft alone is spending $190B on capex this year, up 61%, and still tells investors demand exceeds available capacity. Centralized providers will keep growing, yet they won’t be enough.
Decentralized compute is a release valve, but it has always hit the same wall. You can pay a stranger for work. Proving they did it is the hard part. Blockchains price and settle cheaply. Verification is the bottleneck.
On QMS, miners run solvers on useful problems proposed by clients. The solver hashes its own state as it goes, so a verifier re-runs one chunk instead of the whole computation.
Original Tweet
X | LinkedIn | Medium | Docs
The demand curve keeps steepening. More models, more inference, more work that has to run somewhere. Microsoft alone is spending $190B on capex this year, up 61%, and still tells investors demand exceeds available capacity. Centralized providers will keep growing, yet they won’t be enough.
Decentralized compute is a release valve, but it has always hit the same wall. You can pay a stranger for work. Proving they did it is the hard part. Blockchains price and settle cheaply. Verification is the bottleneck.
On QMS, miners run solvers on useful problems proposed by clients. The solver hashes its own state as it goes, so a verifier re-runs one chunk instead of the whole computation.
Original Tweet
X | LinkedIn | Medium | Docs
❤6👍2
A snapshot of where major ecosystems actually stand on post-quantum migration:
Bitcoin: two numbered BIPs now exist. BIP-360 (quantum-safer outputs, merged Feb 2026) and BIP-361 (a phased legacy-signature sunset, published April 2026). Both are drafts. Nothing is activated, and no migration design has network consensus.
Ethereum: a dedicated EF post-quantum team formed in January. Its public roadmap covers four surfaces: account signatures, validator BLS signatures, data availability, and ZK proof systems. Account abstraction (EIP-8141, under consideration for the Hegota fork) enables gradual user migration. The 2029 infrastructure target is a plan, not a commitment.
Solana: Anza and Firedancer independently converged on Falcon and built initial implementations, the Foundation published a phased readiness roadmap in April, and a Winternitz-based vault primitive has been live in the ecosystem for over two years. Network-wide PQ signatures are not.
Major Ethereum L2s still inherit Ethereum's timeline at the settlement layer. Starknet has the most concrete public plan: a three-phase PQ roadmap published in June, a first mainnet hash migration landing this month, and PQ wallets already deployable through native account abstraction. Its own roadmap concedes that bridge messaging and data availability wait on Ethereum. No L2 has deployed protocol-wide PQ transaction signatures by default.
Original Tweet
X | LinkedIn | Medium | Docs
Bitcoin: two numbered BIPs now exist. BIP-360 (quantum-safer outputs, merged Feb 2026) and BIP-361 (a phased legacy-signature sunset, published April 2026). Both are drafts. Nothing is activated, and no migration design has network consensus.
Ethereum: a dedicated EF post-quantum team formed in January. Its public roadmap covers four surfaces: account signatures, validator BLS signatures, data availability, and ZK proof systems. Account abstraction (EIP-8141, under consideration for the Hegota fork) enables gradual user migration. The 2029 infrastructure target is a plan, not a commitment.
Solana: Anza and Firedancer independently converged on Falcon and built initial implementations, the Foundation published a phased readiness roadmap in April, and a Winternitz-based vault primitive has been live in the ecosystem for over two years. Network-wide PQ signatures are not.
Major Ethereum L2s still inherit Ethereum's timeline at the settlement layer. Starknet has the most concrete public plan: a three-phase PQ roadmap published in June, a first mainnet hash migration landing this month, and PQ wallets already deployable through native account abstraction. Its own roadmap concedes that bridge messaging and data availability wait on Ethereum. No L2 has deployed protocol-wide PQ transaction signatures by default.
Original Tweet
X | LinkedIn | Medium | Docs
👍4❤2