Quantum computing news you don't need a physics degree to follow. A new briefing every morning, with a Hype Check on every big claim. AI-voiced, human-built.
Intel wants to buy Classiq, a quantum software firm, and pair it with its own quantum hardware. The open question is whether quantum can ever make real money at that scale. Today's episode: the proposed deal, plus NIST's post-quantum migration roadmap. https://t.co/ePROsCdL48
IBM just linked two 120-qubit quantum processors with 99% state-transfer fidelity, meaning the signal moved between chips with very little loss. https://t.co/gwuG31C41Z
Five silicon qubits, one shared microwave line, and the control is accurate enough to matter. A RIKEN-led team says it works, which makes wiring the next big scaling problem easier to picture. https://t.co/Di0el9xxal
Google says it cut quantum error rates 40x. Quantinuum says 10x on its Helios machine. Both claim records in the same month. Meanwhile, Bitcoin's quantum-risk clock may now have actual deadlines. Today's episode: is Q-Day getting a schedule? https://t.co/Gs5DRPc90N
Quantinuum says its Helios machine cut errors by a factor of 12, a rare below-threshold result that suggests error correction may be leaving the lab. https://t.co/V0bMxK9fIV
Google is paying $150,000 to whoever can shrink the quantum circuits it takes to run useful code. Quantinuum opened early cloud access to Helios via Oracle, but customer results are still to come. https://t.co/QHcVoglZ6j
Germany has a contract for managed access to an IQM superconducting quantum processor, aimed at aerospace researchers, with workloads hosted in Germany. Why rent quantum muscle instead of owning it? Today's episode: https://t.co/ViYUli4lUS
Quantinuum says its Helios machine suppressed errors by a factor of 12 and that Sandia National Labs calls its Apollo design feasible. Errors are the thing standing between today's shaky quantum chips and useful ones, so that's the number to watch. https://t.co/Suo46JUsOJ
NVIDIA says it simulated 104 qubits of fault-tolerant quantum computing at Jülich. Brian's take: a simulation is a model of a machine, not the machine itself, and the hard part still runs on a supercomputer. https://t.co/N2ZWqwWdT6