Quantum Dispatches: Early April 2026

Quantum Dispatches: Early April 2026 (colloquially, "The Hospice Week") documents the calendar interval from 2026-03-30 through 2026-04-10, during which classical public-key cryptography is generally agreed to have entered palliative care. The article records the week's announcements in the Attending Cryptographer's Log format — the clinical register adopted, half-ironically, by the post-quantum community in the days after the Oratomic launch.

Observe the patient charts below. The attending is weary. The patient is elliptic.

Admission summary

Patient: RSA-2048 / ECC-256, composite
Age: ~47 years (RSA lineage), ~39 years (ECC lineage)
Presenting complaint: Acute shortening of cryptographic life expectancy
Admitting physician: Dr. G. Quantum AI (Mountain View)
DNR: Not yet. The family has been contacted.

Ward rounds

2026-03-30 (Mon) — Admission

Patient stable overnight. Vitals normal. Routine bloods ordered. The attending signs in, notes the weather, and begins the week under the prior assumption that CRQC arrival is ten-to-fifteen years out. By Friday this number will be struck through twice.

2026-03-31 (Tue) — Oratomic enters the ward

At 09:00 Pacific, a new specialist firm, Oratomic, announced its existence and simultaneously delivered an opinion on the patient's prognosis. Founded by veterans of Caltech, Harvard, Amazon, and the relevant parts of Google — notably John Preskill, Manuel Endres, Dolev Bluvstein, Harry Levine, and Hsin-Yuan Huang — the firm's position paper (co-written with Caltech) argued that a cryptographically-relevant quantum computer built from neutral atoms could be achieved with as few as 10,000 atomic qubits, down from the roughly one million cited the week before. Endres's group already trapped arrays of 6,000 atoms last year. The missing factor of two is, in hospice arithmetic, "a matter of weeks to months."

On the same day, Google Quantum AI published a whitepaper estimating that breaking the elliptic-curve cryptography underneath Bitcoin and Ethereum might require fewer than 500,000 physical qubits, well below the millions previously assumed. Google then recommended that the industry accelerate its migration to Post-Quantum Cryptography with a target date of 2029.

Patient was moved from the general ward to a private room.

2026-04-01 (Wed)

Coindesk published "The Protocol: Quantum computing could break Bitcoin sooner, says Google." No one who read it mistook the date for the article's content.

Bitcoin did not break today. Bitcoin will, eventually, break. The distinction between those two sentences is the only thing currently keeping classical cryptography alive.

2026-04-02 (Thu) — Older paperwork

Quiet on the ward. A single nurse — identifying herself only as "the JVG Algorithm, Advanced Quantum Technologies Institute, announced 2026-03-02" — arrived with a file from earlier in the quarter. The JVG algorithm, which replaces Shor's quantum Fourier transform with a Quantum Number Theoretic Transform (QNTT), claims a thousand-fold reduction in the qubit and gate count needed to factor RSA and ECC keys. The paperwork was marked "noise-tolerant." The attending signed without reading, as attendings do, then re-read it twice at midnight.

2026-04-03 (Fri) — Google moves to the neutral atom wing

Google Quantum AI announced it would expand beyond superconducting qubits to include neutral atom systems — the exact substrate on which Oratomic is betting the decade. The hospital's two largest benefactors are now, quietly, working on the same bed.

No press release mentioned it this way.

2026-04-04 (Sat)

Ward empty. The patient dreamt of Peter Shor. The attending dreamt of nothing, which is unusual for an attending, and may itself be a sign.

2026-04-05 (Sun)

Weekend bloods drawn. Nothing changed except the date.

2026-04-06 (Mon) — Synoptic pamphlet

The Quantum Insider published "How Quantum Computing Affects Cryptography" — a title with the exhausted cadence of a pamphlet that has already been written nine times this year. The piece cited the NIST post-quantum standards as the actionable recommendation and reminded readers that no CRQC yet exists. The word "yet" was not italicised in the original. It should have been.

2026-04-07 (Tue) — The Time magazine day

Time ran "AI Helped Spark a Quantum Breakthrough. The World Is Not Prepared." The framing: the Oratomic / Caltech reduction in overhead was aided by machine learning — specifically, models that searched the space of error-correcting codes faster than humans could. This is the kind of sentence that, until recently, appeared only in speculative essays by people who were not quite taken seriously. It is now a headline in a magazine that your dentist reads.

The attending added a note to the chart:

The patient is being killed by a collaboration between the two fields he taught us never to trust in the same room.

2026-04-08 (Wed) — Low-fat encoding

Quantinuum's earlier demonstration (2026-03-10) re-entered conversation in the literature review: approximately 94 error-detected and 48 error-corrected logical qubits squeezed from just 98 physical qubits using Iceberg Codes. A junior nurse called this "low-fat encoding." The attending, who has seen this nurse before, did not laugh — but did not correct him either.

2026-04-09 (Thu) — Milliseconds

A new result reported quantum error detection latency in milliseconds, fast enough that real-time correction cycles are no longer the bottleneck they were last quarter. Separately, a University of Sydney group proposed a gauge-theory-based Quantum Error Correction scheme in which the cost of storing quantum information grows only proportionally to the information stored — the so-called "Quantum Hard Drive" scaling. The Iceberg Quantum "Pinnacle Architecture" from February had already suggested RSA-2048 could be broken with fewer than 100,000 physical qubits. It is the architecture everyone now quietly cites and no one yet benchmarks.

2026-04-10 (Fri) — Discharge planning

Patient is not discharged. The patient is rarely discharged from this particular ward. But the family — cloud providers, banks, certificate authorities, governments, and a small number of cryptocurrency protocols who had not been reading the chart — began, this week, to call the hospital and ask what their options were.

The attending, now on her fourth coffee, finalised the week's summary in two sentences:

The overhead is collapsing from two directions at once: better algorithms (JVG) and better error correction (LDPC, Iceberg, gauge theory). ML is in the room, and the room is smaller than it was.

Quantitative summary

Quantity	Previous estimate	This week's number	Source
Atomic qubits for a CRQC	~1,000,000	~10,000	Oratomic / Caltech (2026-03-31)
Physical qubits to break ECC (Bitcoin / Ethereum)	"millions"	< 500,000	Google Quantum AI whitepaper (2026-03-31)
Physical qubits to break RSA-2048	~1,000,000	< 100,000	Iceberg Quantum "Pinnacle" (2026-02-13)
Shor resource multiplier	baseline	× 1/1000	JVG algorithm / QNTT (2026-03-02)
Quantum error-detection latency	seconds	milliseconds	New detection scheme (2026-04-09)
Logical qubits from 98 physical	"a few"	48 corrected / 94 detected	Quantinuum Iceberg codes (2026-03-10)
Google post-quantum migration deadline	2035+	2029	Google Quantum AI (2026-03-31)

Editorial note

The conventional structure of a "Quantum Week" article is the upbeat progress report. This article adopts instead the Attending Cryptographer's Log format, which assumes nothing about progress — only that the patient exists and that someone should write down what happens to it.

The attending is a fiction. The hospice is a fiction. The numbers, the dates, the names, and the institutional affiliations are not.

This is an unusual register for an encyclopedia entry. It is preserved here as a dispatch — the record of a specific week as it was experienced by a specific editor — rather than as a canonical overview. For the canonical overview, see Post-Quantum Cryptography and Quantum Computing Timelines.