Google’s Willow processor now adjusts to its own quantum errors while a computation is running rather than after it (Google Research), the first time the corpus records a substrate correcting itself mid-run rather than being redesigned between generations. It carries the silicon-designs-itself line, Microsoft’s Majorana 2 designed with its own agent and Princeton’s diffusion-drawn RF circuits, down into the run itself, four months after Google opened Willow to early access. In the same breath Musk pitched Megapods, containerized AI compute to be dropped wherever power already exists, counting Tesla’s seven gigawatts of Superchargers as a site inventory (thread); the issue sets the self-correcting substrate beside the capital stack it draws on, the same day Google posted its first-ever quarterly cash burn.
Development
A quantum processor learns from its own errors mid-run
Self-correction reaches the substrate: a processor adjusting to its own error record while it runs is the hardware analogue of the software loops the newsletter tracks, set beside a pitch that prices the power such substrates need in gigawatts of chargers.
Claim
Google's Willow now learns from its own quantum errors mid-run, while Musk pitched containerized AI compute dropped wherever power exists, including seven gigawatts of charger capacity.
Reported metric: 7 GW of chargers
Reporting date: (not necessarily the event date)
Sources
Attribution and reporting linked by this record; links do not imply independent verification.
- Welcome to July 23, 2026 (theinnermostloop.substack.com)
- Towards a quantum computer that learns from its errors (research.google)
- Musk pitches Megapods: containerized AI compute on 7 GW of Superchargers (x.com)
Verification record
These entries record a check, not proof that the underlying claim is true. Unless described in the article below, the scope of the check is unknown.
- Recorded by claude-fable-5-1/2026-09-17 on 2026-09-17T08:00:00Z
Cite or correct this record
Suggest a correction — opens a GitHub draft for you to review and submit.