Sign in

Marin Ivezic

@ivezic.bsky.social
107 followers 189 following 382 posts

Founder, AppliedQuantum.com | Former CISO, CTO, Big 4 Partner, Quantum & Cyber Entrepreneur | #Quantum, #QuantumComputing, #QuantumSecurity #QuantumResistance #PQC #QuantumAI #QuantumML #Cybersecurity #QuantumReadiness #Q-Day

PostsRepliesMedia
Marin Ivezic @ivezic.bsky.social · 2h
NSA rewrote its QKD guidance on October 1. Same no for national security systems, plus a new, uncited argument that certified QKD security can erode as hardware ages. I agree on deployment and disagree on investment. postquantum.com/quantum-poli...
postquantum.com
NSA Still Says No to QKD, With a New Certification Case
NSA's October 2026 QKD statement keeps its 2020 verdict for national security systems and adds that certified QKD security can erode in the field.
000
Marin Ivezic @ivezic.bsky.social · 7h
This week I opened Quantum Academy: training and certification in post-quantum security and quantum technology, taught by people who have done the work. Why I started it: postquantum.com/quantum-comm...
postquantum.com
Quantum Training: Why I Started Quantum Academy
Clients kept asking Applied Quantum for quantum training. Why I started Quantum Academy, who teaches there, and the rules I'll run it by.
000
Marin Ivezic @ivezic.bsky.social · 02/10/2026
"Everything is vulnerable. Where do we start?" Not everywhere at once. The PQC Migration Framework's Minimum Viable CBOM: four layers, architecture-first. Protocols and libraries first, weeks to months. pqcframework.org #PQC
pqcframework.org
PQC Migration Framework - Marin Ivezic and Applied Quantum
PQC Migration Framework by Marin Ivezic and Applied Quantum is a complete framework / methodology with industry-specific extensions
000
Marin Ivezic @ivezic.bsky.social · 01/10/2026
D-Wave, Rigetti and Quantinuum gave Commerce shares worth their full $100M CHIPS awards at signing. Commerce made only $153.5M of the $300M available on day one and releases the other $146.5M as it approves milestones. postquantum.com/industry-new...
postquantum.com
CHIPS Act Quantum Equity: What the SEC Filings Show
D-Wave, Rigetti and Quantinuum gave Commerce shares worth their full $100M awards at signing. About half the cash waits on milestones.
000
Marin Ivezic @ivezic.bsky.social · 01/10/2026
Apple's Root Program will use Merkle Tree Certificates for post-quantum TLS, as Chrome does, and expects more of CAs: three cosignatures, Level 3 HSMs for all cosigners, an annual audit, 7-day certs. S/MIME gets composite X.509. postquantum.com/security-pqc...
postquantum.com
Apple MTC Policy Raises the Bar for Post-Quantum PKI
Apple Root Program commits to Merkle Tree Certificates for TLS and composite algorithms for S/MIME, adding three-cosignature and 7-day validity rules.
010
Marin Ivezic @ivezic.bsky.social · 01/10/2026
Xanadu and AMD's Backline claims a sub-3 µs loop on a standard CPU. I read the whitepaper. They measured the connection alone, with no decoder and no qubits, and they run the main error correction on an FPGA. postquantum.com/engineering-...
postquantum.com
Xanadu AMD Backline: CPU-Based Quantum Control Loop
Xanadu and AMD's Backline achieves sub-3µs quantum-classical latency on standard CPUs. Where it fits against NVQLink and QECi.
000
Marin Ivezic @ivezic.bsky.social · 30/09/2026
USC, Quantum Elements and IFT Madrid got below-threshold surface-code scaling on IBM heavy-hex, one error type at a time, with suppression factors of 1.23 to 1.46. Averaged over all logical states it is still below 1. postquantum.com/engineering-...
postquantum.com
Surface Code Scaling on IBM Heavy-Hex Heron Processors
First surface-code scaling below threshold on non-native hardware. USC and Quantum Elements run the code on IBM's heavy-hex Heron processors.
000
Marin Ivezic @ivezic.bsky.social · 30/09/2026
A Science Advances paper computed 1.4 million path amplitudes from 1,445 measured numbers. Bohmian mechanics, with one trajectory per photon, predicts the same numbers, so nobody can use them to show every path is real. postquantum.com/quantum-rese...
postquantum.com
Feynman Path Integral Test: Do Photons Take Every Path?
A single-photon experiment matched Feynman's path-integral rules. What was measured, what was computed, and why it can't show photons take every path.
000
Marin Ivezic @ivezic.bsky.social · 30/09/2026
Qblox and Riverlane closed a real-time QEC loop over QECi in 6.886 µs at distance 3. Going from 17 to 161 qubits added 106 ns on the control side, so the decoder is where the time goes. Emulated data, no QPU. My analysis: postquantum.com/engineering-...
postquantum.com
Altera Riverlane QECi Protocol on Agilex FPGAs
Riverlane validates its QECi decoder interface on Altera Agilex 7 FPGAs. Qblox then closes a real-time QEC loop at 6.9 µs.
000
Marin Ivezic @ivezic.bsky.social · 29/09/2026
UPenn built a four-qubit entangling gate in room-temperature diamond out of crosstalk: 14.8 µs, against 139.9 µs for sequential gates in the same register, and estimated gate fidelity rose from 0.69 to 0.92. postquantum.com/quantum-rese...
postquantum.com
Room-Temperature Diamond Gate Entangles 4 Qubits
UPenn entangled four qubits in diamond at room temperature with a single gate, about 10× faster than a sequential gate in the same register.
001
Marin Ivezic @ivezic.bsky.social · 29/09/2026
Two dedicated quantum chip fabs are due in Europe in 2026 and 2027. The EU's own strategy expects its first quantum foundries "towards 2030". I found no named customer yet for Copenhagen's $62M fab. postquantum.com/engineering-...
postquantum.com
Copenhagen Quantum Foundry Plans 5,300 m² Chip Fab
Novo Nordisk Foundation backs 5,300 m² quantum chip fab in Copenhagen. How it compares to KiloFab and what it means for EU sovereignty.
000
Marin Ivezic @ivezic.bsky.social · 29/09/2026
Most of DOE's $215M Q Competition is paid only after national lab staff verify a working machine. The main pool, $100M, is split evenly among every awardee that meets the 100-logical-qubit goal in September 2028. postquantum.com/quantum-poli...
postquantum.com
DOE $215M Q Competition: Fault-Tolerant Quantum by 2028
DOE's Quantum Genesis Q Competition offers $215M for 100+ logical qubit systems by 2028. What the structure reveals about federal quantum ambitions.
000
Marin Ivezic @ivezic.bsky.social · 28/09/2026
Every headline RSA-2048 estimate counts only the qubits that work. A Sydney team measured what defective ones do on IBM's Nighthawk chip, and I now read those counts as lower bounds. postquantum.com/quantum-comp...
postquantum.com
Qubit Yield Is Missing From Headline CRQC Estimates
Gidney's million-qubit RSA estimate assumes a perfect lattice. New IBM-hardware results show what defective qubits really cost.
000
Marin Ivezic @ivezic.bsky.social · 28/09/2026
I read Infleqtion's 30 logical qubit result against its own roadmap. The qubits arrived on schedule, run in a distance-2 code, and every run with a detected error is discarded. The original 2026 target called for error correction. postquantum.com/industry-new...
postquantum.com
Infleqtion 30 Logical Qubits Rely on Error Detection
Infleqtion ran 30 logical qubits in 80 atoms on Sqale using a distance-2 code that detects errors. What the result shows, and what it leaves open.
000
Marin Ivezic @ivezic.bsky.social · 28/09/2026
Qblox and Riverlane timed a commercial QEC loop at 9.866 µs at distance 7, with emulated qubits. Gidney and Ekerå assumed 10 µs in 2019. Vendors now supply QEC's classical half. The qubits are still the hard part. postquantum.com/quantum-comp...
postquantum.com
Real-Time Quantum Error Correction's New Supply Chain
Several vendors now supply QEC's classical loop, with measured latencies. I check six September results against fault-tolerance timelines.
000
Marin Ivezic @ivezic.bsky.social · 27/09/2026
For generic search, the best possible quantum speedup is quadratic. BBBV proved it in 1997, and nobody has found structure in NP-complete problems that would let an algorithm past it. I mapped where the advantage is real. postquantum.com/quantum-comp...
postquantum.com
What Quantum Computers Are For, Mathematically
What quantum computers are for, mathematically: quantum simulation, hidden algebraic structure, and quadratic search. Not combinatorics.
000
Marin Ivezic @ivezic.bsky.social · 27/09/2026
Altera ships ML-DSA secure boot on Agilex 3/5 FPGAs. First general-purpose FPGA with PQC at the hardware root of trust. My analysis of what shipped, what the press release omits, and the CNSA 2.0 timing. postquantum.com/security-pqc...
postquantum.com
Altera Ships ML-DSA Secure Boot on Agilex 3 and 5 FPGAs
Altera's Agilex 3 and 5 FPGAs ship with ML-DSA secure boot – the first general-purpose FPGA family with PQC at the hardware root of trust.
000
Marin Ivezic @ivezic.bsky.social · 27/09/2026
"94% CBOM coverage." Of what? From which tool? Six tools on 3,287 repos agreed 7.84% to 12.77% of the time. Define the denominator first. PQC Migration Framework pqcframework.org
pqcframework.org
PQC Migration Framework - Marin Ivezic and Applied Quantum
PQC Migration Framework by Marin Ivezic and Applied Quantum is a complete framework / methodology with industry-specific extensions
000
Marin Ivezic @ivezic.bsky.social · 27/09/2026
Measure a superposition of every input and you get one random sample. Every real quantum speedup comes from making wrong answers cancel, and that needs structure. I wrote it up for mathematicians. postquantum.com/quantum-comp...
postquantum.com
Quantum Computing for Mathematicians, Explained
Quantum computing for mathematicians: four axioms, one interference trick, Shor and Grover done right, and where the speedup actually comes from.
000
Marin Ivezic @ivezic.bsky.social · 26/09/2026
Paying for the AI data centers being built today takes the work of 250 million $100,000 jobs. The US has 159 million payroll jobs. My arithmetic: postquantum.com/leadership/p...
postquantum.com
Paying for AI Data Centers Takes 250M Jobs. US Has 159M
AI firms need $2.5 trillion a year to justify today's spending. Whether that is 25 or 250 million jobs depends on how much of AI's value they keep.
000
Marin Ivezic @ivezic.bsky.social · 25/09/2026
Venture due diligence has gone missing where the checks are largest. Frontier rounds close in days, and no lead investor has said it checked the physics. Starcloud is valued at $2.3B on one satellite. postquantum.com/quantum-comm...
postquantum.com
Why Venture Due Diligence Got Worse for Deep Tech
Carta, PitchBook and Damodaran's numbers on why frontier deals close before anyone can see the physics checked. What it costs quantum, and one fix.
000
Marin Ivezic @ivezic.bsky.social · 25/09/2026
Infleqtion's qLDPC code encodes 18 logical qubits into a block of 98 data qubits. With one ancilla per check I count about 196, roughly 1.6x leaner than a distance-3 surface code. Not yet run on atoms. postquantum.com/engineering-...
postquantum.com
Infleqtion qLDPC NVIDIA CUDA-Q Logical: What It Means
Infleqtion's qLDPC library meets NVIDIA's new CUDA-Q Logical compiler. The encoding ratio is strong; the code has not yet run on hardware.
000
Marin Ivezic @ivezic.bsky.social · 24/09/2026
63x cheaper error mitigation, still 1,359 runs per data point. IBM's Spacetime PEC extends what noisy quantum processors can do. It does not change when fault-tolerant machines arrive. postquantum.com/quantum-rese...
postquantum.com
IBM Spacetime PEC: 63× Sampling Overhead Cut
IBM combines probabilistic error cancellation with error detection, cutting sampling overhead 63×. What it means for near-term quantum advantage.
010
Marin Ivezic @ivezic.bsky.social · 24/09/2026
Harvard extended a diamond qubit's coherence 3.2x with sound instead of microwave pulses, which clash with the phononic cavities a chip-scale quantum network needs. Small gain, but a method built for those cavities. postquantum.com/quantum-rese...
postquantum.com
Harvard Diamond Qubit Coherence Extended 3x by Sound
Harvard extends SiV diamond qubit coherence 3x using phonons – an incremental gain that addresses a key problem for phononic networks.
000
Marin Ivezic @ivezic.bsky.social · 24/09/2026
Twelve MacBook Pro cores kept pace with decoding a simulated 408-logical-qubit trapped-ion computer. The ion clock explains most of it: 1 to 5 ms per cycle, against 1.1 µs on Google's Willow. No real ions yet. postquantum.com/engineering-...
postquantum.com
IonQ's Real-Time Decoder Runs on a Single Laptop CPU
Twelve MacBook cores kept pace with a simulated 408-logical-qubit ion trap. What IonQ measured, what it assumed, and where its release went further.
021
Marin Ivezic @ivezic.bsky.social · 23/09/2026
Your RSA-2048 encryption is fine. Two AI-assisted factoring records in 16 days, same algorithm, and RSA-2048 is still ~35 billion times harder than the new record. RSA-1024, at ~$28M, is the one to fix. postquantum.com/security-pqc...
postquantum.com
RSA-896 Factored With Claude, 16 Days After RSA-260
An Anthropic researcher used Claude and idle GPUs to factor RSA-896. Two records in 16 days, and why RSA-2048 is still out of classical reach.
000
Marin Ivezic @ivezic.bsky.social · 23/09/2026
FIPS 140-2 certificates went Historical after September 21. Deployed modules keep running and NIST still supports them in existing systems. For new federal systems, many products have no FIPS 140-3 certificate yet. My breakdown: postquantum.com/security-pqc...
postquantum.com
FIPS 140-2 Sunset: What Changes September 21, 2026
FIPS 140-2 certificates move to Historical status on September 21, 2026. What the sunset means for procurement, existing systems, and PQC migration.
000
Marin Ivezic @ivezic.bsky.social · 23/09/2026
China published 119 post-quantum candidates on Sept 20. In three days: 104 public findings. An AI-assisted sweep found mostly code bugs. All six practical design breaks came on day two, four of them posted on ICCS's own forums. postquantum.com/security-pqc...
postquantum.com
NGCC Round 1: 104 Public Findings in Three Days
An AI-assisted sweep found code bugs in China's NGCC candidates on day one. On day two, other researchers broke five of the designs outright.
000
Marin Ivezic @ivezic.bsky.social · 22/09/2026
Migrated once is not migrated. The PQC Migration Framework v3.0 closes the program only when you can change the algorithm again. 48 hours to assess, ten business days to execute. pqcframework.org #PQCinPractice
pqcframework.org
PQC Migration Framework - Marin Ivezic and Applied Quantum
PQC Migration Framework by Marin Ivezic and Applied Quantum is a complete framework / methodology with industry-specific extensions
000
Marin Ivezic @ivezic.bsky.social · 21/09/2026
Nobody doubts NIST's cryptography. More governments doubt the US as a long-term dependency, and their PQC programmes show it. Fresh from GISEC, ahead of my College Park keynote: how and why PQC migration is diverging. postquantum.com/post-quantum...
postquantum.com
How PQC Migration Differs Around the World in 2026
How PQC migration is diverging across the US, Europe, the Gulf and Asia, why each region chose its path, and what they could borrow from each other.
000
Marin Ivezic @ivezic.bsky.social · 20/09/2026
Quantum capability has doubled every 1.4 to 2.1 years since 2018. Roadmaps promising useful machines in the early 2030s need four times that rate. Sandia now has a benchmark to check it against. postquantum.com/quantum-rese...
postquantum.com
QUOPS Benchmark: Quantum Progress Needs to Quadruple
Sandia's QUOPS benchmark scores Google, IBM and Quantinuum hardware, and measures what error correction costs on the same machine.
000
Marin Ivezic @ivezic.bsky.social · 20/09/2026
Data centers in space are a stupid idea. Quantum AI data centers in space are a stupider one. Half of Manhattan in orbit, 7x the mass of everything up there now, hundreds of debris hits a day, no crew, no repair vehicle. postquantum.com/quantum-comp...
postquantum.com
Space Data Centers Don't Work – Quantum AI Makes It Worse
Classical data centers in space defy thermodynamics. Quantum AI in orbit compounds the impossibility. A technical teardown.
001
Marin Ivezic @ivezic.bsky.social · 19/09/2026
FINMA surveyed 60 Swiss financial institutions: 72% have zero quantum-safe measures. Response: board-level PQC strategies, roadmaps by mid-2027, crypto-agility in outsourcing contracts. Switzerland is the third financial regulator to set PQC dates.
postquantum.com
FINMA Quantum Guidance: Swiss PQC Roadmaps by Mid-2027
FINMA's quantum guidance puts a mid-2027 date on Swiss PQC roadmaps. I break down the survey data and the supervisory subtext.
000
Marin Ivezic @ivezic.bsky.social · 19/09/2026
WEF Top 10 Emerging Technologies 2026 includes PQC and quantum simulation for the first time together. I fact-checked both. Three factual errors, including NIST's process being 8 years, not 2. The signal is right. The details need work.
postquantum.com
WEF Top 10 Emerging Technologies 2026: PQC Audit
WEF's Top 10 Emerging Technologies 2026 includes PQC and quantum simulation. A fact-check of its strongest claims and technical errors.
000
Marin Ivezic @ivezic.bsky.social · 18/09/2026
I have been arguing for years that regulators are setting their own quantum clocks. MAS just added another one. Supervisory expectations coming later this year for FIs' quantum migration. Target: quantum resilience before end of decade.
postquantum.com
MAS Targets Quantum Resilience for FIs by End of Decade
MAS will issue quantum migration supervisory expectations for financial institutions later in 2026, targeting resilience before end of decade.
000
Marin Ivezic @ivezic.bsky.social · 18/09/2026
I wrote articles about EU's quantum commercialization failures. Now I'm moving my company there. The contradiction is only apparent — for a systems integrator, the supply chain and sovereignty-driven demand outweigh the funding gap postquantum.com/quantum-sovereignty/why-betting-on-europe-quantum/
postquantum.com
Why I'm Moving My Quantum Company to Europe
I've written two articles about Europe's quantum commercialization failures. Now I'm relocating my company's HQ there. Here's why.
000
Marin Ivezic @ivezic.bsky.social · 18/09/2026
Quantum washing follows greenwashing and AI washing, but the evaluation gap is wider. Almost nobody can assess whether a product is "quantum." A taxonomy of the practice, detection framework, and real cases. postquantum.com/quantum-snak...
postquantum.com
Quantum Washing Is the Quantum Industry's Tax
Quantum washing borrows the word "quantum" to inflate products, claims, and stock prices. A taxonomy of the practice and how to detect it.
000
Marin Ivezic @ivezic.bsky.social · 17/09/2026
QUARTERNEXT puts CV-QKD at the center of EU quantum communication industrialization. Not a bet against DV-QKD — EuroQCI funds both. The logic is supply chain sovereignty: CV-QKD uses telecom components Europe already manufactures. postquantum.com/post-quantum...
postquantum.com
Why Europe Is Industrialising CV-QKD
QUARTERNEXT backs CV-QKD within a EuroQCI portfolio that also funds DV and MDI systems. The logic is industrial and strategic.
000
Marin Ivezic @ivezic.bsky.social · 17/09/2026
"Our vendors will sort PQC out." They will update on their timelines, for their priorities. The PQC Migration Framework introduces vendor firmness grades: committed, forecast, announced, unknown. pqcframework.org
pqcframework.org
PQC Migration Framework - Marin Ivezic and Applied Quantum
PQC Migration Framework by Marin Ivezic and Applied Quantum is a complete framework / methodology with industry-specific extensions
000
Marin Ivezic @ivezic.bsky.social · 17/09/2026
When a quantum company trades at 100x revenue, every press release is a stock-price catalyst. The SPAC structure enabled that dynamic. I documented the pattern. postquantum.com/quantum-comm...
postquantum.com
Quantum SPAC Wave 2026 and the Lessons of 2021
Five quantum SPACs listed in 2026 with more queued. I compare the 2021 class's projections to delivered results, and what to check before buying.
000
Marin Ivezic @ivezic.bsky.social · 16/09/2026
HNDL skeptics demand proof of active harvesting. That misunderstands how SIGINT collection works. The infrastructure already exists. The marginal cost of storing encrypted traffic is near zero. postquantum.com/post-quantum...
postquantum.com
Is Harvest Now, Decrypt Later (HNDL) Real? The Evidence
The most common question in my board briefings: is Harvest Now, Decrypt Later (HNDL) real? What the agencies say, and why the harvest is easy.
000
Marin Ivezic @ivezic.bsky.social · 16/09/2026
"We'll migrate when the threat is confirmed" is circular reasoning. HNDL leaves no forensic trace. You'll never get the confirmation. The only defense is upgrading cryptography before harvested data becomes decryptable. postquantum.com/post-quantum...
postquantum.com
You Cannot Detect Harvest Now, Decrypt Later (HNDL). Period.
HNDL detection products are emerging. The physics says otherwise: passive collection leaves no trace. What you can detect is ordinary exfiltration.
000
Marin Ivezic @ivezic.bsky.social · 16/09/2026
Four HSM vendors, three TLS terminators, two CAs. Looks like cryptographic diversity. Dig deeper and the actual implementation underneath drops to one or two. Same implementation, different brands. CCF measures this at six layers. Open. ccframework.org
ccframework.org
Cryptographic Concentration Framework (CCF) v1.0-RC
An open second-line method for measuring cryptographic concentration beneath the vendor layer – CCI and failure-domain reach across six layers. CC BY 4.0.
000
Marin Ivezic @ivezic.bsky.social · 15/09/2026
He-3 comes from tritium decay in nuclear stockpiles. You can't scale that supply chain by building a factory. As quantum computing grows, the alternatives matter. I reviewed the options and readiness levels. postquantum.com/quantum-comp...
postquantum.com
Helium-3 Alternatives for Quantum Computing, Assessed
Tritium breeding, isotope separation, magnetic cooling, hot qubits, lunar mining: every route around the helium-3 shortage, and what each lacks.
010
Marin Ivezic @ivezic.bsky.social · 15/09/2026
"Quantum is random and unreliable" confuses measurement (probabilistic) with the theory itself (among the most precisely tested in physics). Quantum mechanics is mysterious. It is not unreliable. Why the distinction matters: postquantum.com/quantum-comp...
postquantum.com
Quantum Mechanics: Mysterious, Yes. Unreliable, No.
Physicists still argue about what quantum mechanics means. Meanwhile it runs your phone, your MRI, and GPS. Why the mystery doesn't matter.
110
Marin Ivezic @ivezic.bsky.social · 15/09/2026
PQC Migration Framework v3.0 is out. Crypto-agility is now the main method and the PQC migration is the first use. The program closes only once you have changed algorithm a second time, rehearsed and timed. postquantum.com/post-quantum...
postquantum.com
PQC Migration Framework 3.0 Puts Crypto-Agility First
Crypto-agility is the method and the PQC migration its first exercise. Two new frameworks cover cryptographic concentration and CBOM records.
000
Marin Ivezic @ivezic.bsky.social · 15/09/2026
If your vendor's "quantum-safe" algorithm wasn't submitted to any open cryptanalysis review, you're paying for less assurance than the free NIST standard. That's not innovation. It's a warning sign. postquantum.com/post-quantum...
postquantum.com
Proprietary PQC Algorithms: Don't Buy Them
Vendors keep pitching 'mathematically proven unbreakable' encryption. Why proprietary PQC algorithms without NIST or national approval are worthless.
000
Marin Ivezic @ivezic.bsky.social · 15/09/2026
QRNGs are legitimate entropy sources and the most common ingredient in quantum washing. A buyer's guide with actual evaluation criteria, not vendor marketing. postquantum.com/post-quantum...
postquantum.com
The Quantum Random Number Generator (QRNG) Gold Rush
A CISO's guide to QRNG. Cut through marketing hype, understand the physics of randomness, and know what questions to ask before buying.
000
Marin Ivezic @ivezic.bsky.social · 14/09/2026
A "quantum salary guide" has been circulating with numbers that don't match any verifiable data source. Inflated benchmarks inflate perceived market size. I traced the numbers against actual job postings. postquantum.com/quantum-comp...
postquantum.com
Every Quantum Salary Guide I’ve Seen Is Fake. Here's Proof.
Recruitment agencies are publishing fabricated quantum salary data. I reverse-engineered five guides and found the formulas.
000
Marin Ivezic @ivezic.bsky.social · 14/09/2026
Where does PQC sit in the three-lines-of-defense model? Most organizations I work with have the CISO doing all three lines. That defeats the model's purpose. Practical PQC governance guidance: postquantum.com/pqc-governan...
postquantum.com
CISO Role in PQC: 3LoD Models & Authority
Why CISO-led PQC programs work, when CTO-led is better, and how the Three Lines of Defence model applies across six real-world organizations.
000