The Quantum Doom Clock is Ticking, March 2025

GM frens, this is the first edition of the Quantum Doom Clock newsletter, with Colton Dillion and Rick Carback, the founders of Quip Network: the only protocol that defends your crypto from quantum computer hackers with no need to bridge funds or change out your keys. We put together this site and newsletter because the clock is ticking toward the day when your assets can be stolen by hackers with sufficiently powerful quantum computers.

For many years, academics, tech founders, and Fortune 500 CEOs have often repeated that useful quantum computers are ten to twenty years away. But those tunes are changing and the people who should be most in-the-know are universally adjusting their timelines:

• Bill Gates - "There's a possibility quantum computing will become useful in 3-5 years."
  (Video Clip)
• Sundar Pichai, Google CEO - "In a 5-10 year timeframe, I think you will start [to see] practically useful quantum computers."
  (Video)
• Hartmut Neven, Google VP of Quantum AI Lab - "We're optimistic that within five years we'll see real-world applications that are possible only on quantum computers."
• Satya Nadella, Microsoft CEO - "We believe this breakthrough will allow us to create a truly meaningful quantum computer not in decades, as some have predicted, but in years."
• Arvind Krishna, IBM CEO - "We feel over the next three, four, five years — I give myself till the end of the decade — we will see something remarkable happen on that front and I’m really happy where our team is."

The odd men out here are Mark Zuckerberg at Meta, who readily admitted he doesn't know much on the Joe Rogan Experience, and Jensen Huang at Nvidia, who is sticking to the old estimates---and really, really wants you to keep buying his GPUs.

The takeaway can be summed up with this gem from a long-time investor in this space, Peter Barret: "Useful quantum computing is inevitable—and increasingly imminent."

Following Google's Willow announcement late last year and their results in Nature, the rest of big tech has jumped on the major result bandwagon alongside some newer players:

• Amazon's Ocelot - with 5 "cat" (data) qubits, a buffer circuit for each for data stabilization, and 4 additional error correcting qubits. Published in Nature.

• Microsoft's Majorana 1 - with 8 qubits using a topological superconducting material, claiming a "new state of matter" that only previously existed in theory. Published in Nature.

• IQM Released a 50 qubit machine in partnership with the VTT Research Center in Finland. This follows a 5 qubit release in 2021 and a 20 qubit machine in 2023. We will now add this to our projections since there is now enough data for them.

• Quantum Machines, which makes quantum control systems, announced a $170M capital raise.

• QuEra, which has a 256 qubit processor, announced another $230M capital raise.

Don’t forget, we also have other major producers like Intel, IBM, Riken, MIT, Japan, and China, among many smaller companies, publishing regular progress and major public quantum computer results each year. We are tracking over 20 different manufacturers in our dataset.

It is worthy to note that the quantum computing revolution will be much like the AI revolution in that the key barriers to progress are theoretical rather than procedural. Once we have a highly performant logical qubit composed of repeatable physical qubit arrays, the manufacturing processes to pack those arrays into 5nm size are already well understood and operating at scale. For example:

• PsiQuantum is making millions of light-based quantum chips with a new manufacturing technique. "scaling to millions of qubits will be relatively straightforward."

• EeroQ claims enhanced readout speeds converting qubit measurements to classical bit outputs, which would significantly reduce our current 8-12 hr estimate for breaking keys.

• Oxford used entanglement to transmit data, enabling larger modular systems to be built. Hooking together chips with quantum teleportation means you don’t have to be limited by the wafer sizes available at the factory.

• UNSW researchers demonstrated an "antimony" qubit, which requires 8 errors to flip a qubit. Error correcting with these brings the physical qubit requirement down to 10s of thousands of qubits for cracking cryptographic keys.

Majorana, Ocelot, and Willow all currently claim better than 10^-3 error rates. The trend is now converging toward 10^-4, so we are updating our projection accordingly. Specifically, we're adding a 0 to the `p` value and rounding the `L` logical error rate down to `1e-16`, which gives us a new code distance of 15. Instead of September 2028, the doom clock will now read:

Have a question? Just e-mail us at team at quantumdoomclock dot com. Below are the top questions we have received since our last update.

Microsoft previously retracted their topoconductor research papers, is the Majorana announcement the real thing?

The paper itself has a caveat by the editors saying there is no "evidence for the presence of Majorana zero modes in the reported devices."

We believe it is real, though, because of comments by one of the authors, Chetan Nayak, explaining that the zero modes and topological qubit claims are based on progress that occurred in the year following the paper submission.

You can learn more via Scott Aaronson's blog post on the matter.

Could Bybit be covering for a quantum computer hack?

No. The announcement of Bybit's $1.5B loss to North Korea's Lazarus group had a notoriously unclear announcement about how they were hacked, but that has since been clarified. The hackers took control of the multisig wallet interface provided by the Gnosis Safe app.