Public-benefit review / 2026

What did the public get?

This is not a fundraising document. It is an accounting of a year, including the programs that did not work.

A draft summary of our annual public-benefit review. It records what was used, by whom, and where we fell short. The full review is shared with our community advisory group before publication.

Draft · October 2026
FY2026
Corrections welcome

4,100 students reached

across 34 public schools

14 reproduction reports

published since 2023

12 public agencies advised

on quantum procurement

1,800 lab visits

free, no registration

5 public tools maintained

with hardware-free tests

Metrics are necessary and insufficient. A visit count does not tell us whether a student learned anything, and a paper count does not tell us whether anyone used it. The case studies below are our attempt to look closer.

01 / What worked

Three things that travelled.

The decoder reproduced, then corrected

An independent group reproduced our central surface-code result and found a discrepancy in a secondary latency claim. We published an erratum and released the missing specification. The exchange is now a template for our reproduction reports.

Read the note

Three water districts kept the instruments

Our portable magnetometers left the lab, survived a thunderstorm, broke twice, and are still collecting weekly transects. District staff now calibrate the instruments themselves — which was always the point.

Sensing program

A classroom study changed our lessons

A six-lesson unit tested with 412 students showed that the popular “both 0 and 1” framing creates persistent misconceptions. We rewrote the materials and released the evidence, including the parts that did not improve.

Classroom paper

02 / What didn't

The failures we can name.

We closed a materials program

A quantum-chemistry project on a promising catalyst family failed to produce a defensible resource estimate after 18 months. We published the negative results, reassigned the team, and wrote the failure into this review.

We abandoned our patent portfolio

The board reviewed our four provisional patents in 2024 and concluded they were inconsistent with the lab's public-benefit charter. We dedicated the underlying work to the public and stopped filing.

A field deployment failed

One of four magnetometers failed calibration after water ingress. We refunded the partner district's deployment cost, published the failure log, and redesigned the enclosure.

03 / Next year

What we owe next.

A modular testbed, in public

Publish the full architecture, thermal budgets, and failure modes for the modular path to 10,000 logical qubits — including the assumptions most likely to break.

Benchmark v2, with users

Rebuild QuBench around tasks nominated by public agencies and university labs, not just vendors, and publish every result, including the ones that flatter us.

Nine more districts

Expand the teacher fellowship to nine additional public-school districts, with a paid advisory group of teachers designing the next edition.

An honest cost of ownership

Publish what it actually costs to run a small quantum lab for a year, including the maintenance nobody budgets for, so other public institutions can plan.

Corrections

If we got this wrong, say so.

This review is a draft. We will publish a correction for any factual error and credit the person who raised it, unless they prefer otherwise.

Send a correction Support the work