SIGNIFICANCE

BOARD ten-results / as of 2026-08-07T00:00:00Z

Board — as of 2026-08-07T00:00:00Z

The ten results: what is actually recorded

One row per result in the August 2026 OpenAI release, answering the question people keep asking — are these real? — in the only terms this project can answer it: what has been checked, what has not, and as of when. Nine of the ten rows are empty because nobody here has done that work yet. An empty row means nobody has looked, not that nothing is there.

Plain-language restatement · Significance editor · 2026-08-07

10 rows · 1 recorded

  1. The unit group L_F2(1,2)× of the binary Leavitt algebra is not sofic.

    Quoted from the source · OpenAI · 2026-08-01, p.78

    Manuscript
    Open link
    Lean target
    Open link
    Checked
    The manuscript file was downloaded and hashed. The published Lean code was rebuilt at a pinned commit and its axiom closure recorded, both logs hashed. Checks on artifacts, not on the argument.
    Not checked
    Nobody has traced the manuscript's Theorem 1.1 to the statement the Lean code contains; the formal target is a broader existence statement. No mathematician's assessment is recorded.
    As of
    2026-08-02T19:33:28Z

    Plain-language restatement · Significance editor · 2026-08-07

    Copy this row into a thread

    Copy and paste this summary.

    A non-sofic group
    
    The unit group L_F2(1,2)× of the binary Leavitt algebra is not sofic.
    
    Checked: The manuscript file was downloaded and hashed. The published Lean
    code was rebuilt at a pinned commit and its axiom closure recorded, both
    logs hashed. Checks on artifacts, not on the argument.
    
    Not checked: Nobody has traced the manuscript's Theorem 1.1 to the statement
    the Lean code contains; the formal target is a broader existence statement.
    No mathematician's assessment is recorded.
    
    As of 2026-08-02T19:33:28Z
    Board: https://hjyuh.github.io/significance//boards/ten-results/
    Significance records evidence. It does not judge the mathematics.
    
  2. [FILL: verify from release]

    Nobody here has researched this result yet. There is no claim, no link and no status recorded for it — an empty row means nobody has looked, not that nothing is there.

  3. [FILL: verify from release]

    Nobody here has researched this result yet. There is no claim, no link and no status recorded for it — an empty row means nobody has looked, not that nothing is there.

  4. [FILL: verify from release]

    Nobody here has researched this result yet. There is no claim, no link and no status recorded for it — an empty row means nobody has looked, not that nothing is there.

  5. [FILL: verify from release]

    Nobody here has researched this result yet. There is no claim, no link and no status recorded for it — an empty row means nobody has looked, not that nothing is there.

  6. [FILL: verify from release]

    Nobody here has researched this result yet. There is no claim, no link and no status recorded for it — an empty row means nobody has looked, not that nothing is there.

  7. [FILL: verify from release]

    Nobody here has researched this result yet. There is no claim, no link and no status recorded for it — an empty row means nobody has looked, not that nothing is there.

  8. [FILL: verify from release]

    Nobody here has researched this result yet. There is no claim, no link and no status recorded for it — an empty row means nobody has looked, not that nothing is there.

  9. [FILL: verify from release]

    Nobody here has researched this result yet. There is no claim, no link and no status recorded for it — an empty row means nobody has looked, not that nothing is there.

  10. [FILL: verify from release]

    Nobody here has researched this result yet. There is no claim, no link and no status recorded for it — an empty row means nobody has looked, not that nothing is there.

What this board does not establish

  • No row says whether a result is right. "Checked" describes artifacts — files hashed, code rebuilt — not arguments read.
  • A row with more text in it is not doing better than a row with less. The rows are not ranked and are not ordered by anything but their place in the release.
  • An empty row means nobody here has looked at that result yet, and nothing more than that.
  • The board's own as-of date is when the board was gone over; each row carries its own, which may be older.