← All problems

The unknotting problem for surfaces in the 4-sphere

For each integer g≥0g\ge 0, determine whether there is an algorithm that, given a closed, connected, orientable, locally flat PL surface F⊂S4F\subset S^4 of genus gg, decides whether FF is TOP unknotted, respectively PL unknotted. The cited source proves that no such algorithm exists for surfaces of genus at least 1818 in the TOP category, and no such algorithm exists for surfaces of genus at least 1919 in the PL category.

References

Source: arXiv

Progress summary

Refreshed
Open · progress reported

A new unrefereed preprint proves strong undecidability results for surface unknotting, but does not settle every genus.

The problem concerns algorithmically recognizing whether surfaces in S4S^4 are unknotted. Cameron Gordon announced smoothly unknotted non-orientable surfaces that are algorithmically unrecognizable in a talk on February 4, 2025.

Recent undecidability results

A preprint by Cameron Gordon claims undecidability at explicit genus thresholds and constructs surfaces whose groups have unsolvable word problem or contain every recursively presented group. The result is a substantial advance, but it does not determine the behavior for every genus and has not been independently verified.

Current status (as of October 2026): Undecidability is claimed at explicit genus thresholds, while the full genus-by-genus unknotting problem remains open and the new preprint is unverified.

Sources

Solutions 0

No solutions have been posted yet.