Open on the site (OPEN; page last edited 21 June 2026). The site's remarks credit the bounds $k^{1+c}<g(k)\le\exp((1+o(1))k)$ to Ecklund, Erdős and Selfridge, and the lower-bound record $g(k)\gg\exp(c(\log k)^2)$ to Konyagin. Ecklund, Erdős and Selfridge write that $g(k)<L_k=\operatorname{lcm}(1,\ldots,k)$ "seems to hold for all $k$" (their conjecture on p. 649); their own table gives $g(k)>L_k$ at $k=2$, $3$ and $6$, so the question is whether $g(k)<L_k$ for all large $k$. The standing in the frontmatter derives from the claim pages: the only claim is the pending partial claim Yang's eventual lcm bound, a manuscript of September 2026 with a Lean development, produced with GPT-6 Astra and GPT-5.6 Sol, proving $g(k)<L_k$ for every sufficiently large $k$; it does not estimate $g(k)$, no reviewer has accepted it, this corpus has not built its Lean, and the site's label is unchanged, so the problem stands open with a pending partial claim.