Dixmier’s problem about spectra of C*-algebras

Let AA and BB be simple separable C∗C^*-algebras, and let A^\widehat{A} and B^\widehat{B} denote their spectra, namely the spaces of unitary-equivalence classes of irreducible unitary representations equipped with their relevant Borel structures. Is there always a Borel-definable bijection between A^\widehat{A} and B^\widehat{B}, where a function between spectra is Borel-definable when it is induced by a Borel map between standard Borel spaces of unitary representations?

References

Primary source

arXiv

Additional references

Progress summary

Refreshed
Claimed solved

A September 2026 unrefereed preprint claims to settle Dixmier’s long-standing spectrum problem, but the result has not yet been independently verified.

Dixmier posed the problem in 1967: whether the relevant Borel structure on spectra of simple separable C∗C^*-algebras is uniquely determined in the nonstandard case. It is also connected with a question of Simon Thomas.

Known results

  • A 2009 dichotomy shows that the pure-state equivalence relation is either smooth or contains a continuous reduction of the $$-equivalence relation, ruling out classification by countable structures in the nonsmooth case.
  • Elliott proved that the corresponding Borel structures of simple separable AF algebras are isomorphic.

September 22, 2026 claimed solution

Martino Lupini’s preprint claims nonexistence of a Borel-definable injection for a specific nuclear/non-nuclear pair, while proving contrasting positive results for separable nuclear non-type-I algebras. If correct, this resolves Dixmier’s 1967 problem and answers Thomas’s related question; the claim is presently unverified.

Current status (as of September 2026): A preprint claims a complete resolution, but independent verification is not recorded, so the problem is not established as solved.

Sources

Solutions 0

No solutions have been posted yet.