Burr and Erdős ([BuEr76], p. 257) print the conjecture with no range on $n$, and the site's wording keeps $n$ free. Read as the site words it, it includes the one-vertex tree $K_1$, where it is false: the host $K_{2n-2}$ is $K_0$, which has no vertex and so contains no copy of $K_1$, and $R(K_1)=1>0=2n-2$. Hua Xu noted this in the site's discussion on 1 May 2026, and two Lean developments in Boris Alexeev's repository, both of 26 August 2026, prove the same failure (Erdos547.not_erdos_547 in Erdos547.lean and Erdos547b.not_literalErdos547 in Erdos547b.lean). It is the only failing instance: every tree on two, three or four vertices is a path or a star, Theorem 1 of [GeGy67] gives $R(P_n)=n-1+\lfloor n/2\rfloor$, that is $2$, $3$ and $5$, and the two-color star value $R(K_{1,3})=6$ completes $n=4$, each at most $2n-2$. The site reads the question as the bound for every $n$: its label DECIDABLE ("Resolved up to a finite check") records that the large-$n$ bound is proved and a finite range of small $n$ remains, the reading its editor gave in the discussion on 11 September 2025 when changing the label from solved to decidable because "the question was for all $n$"; the commentary credits the large-$n$ bound to the Erdős--Sós implication and to Zhao [Zh11]. The site has not addressed $n=1$, and its label and commentary (last edited 18 January 2026) predate the proof of #548 (3 September 2026) and the two requests in the discussion (6 and 17 September 2026) to relabel the problem proved. The corrected Statement excludes exactly $n=1$, the one value at which, because of its size, no host can meet the conclusion; it is the corpus's correction, not the site's, and the formal-conjectures statement file, which counts with the site, states the bound for $n\geq2$ as well (Formalization, below). Under the site's wording the answer is: false at $n=1$ and true for every $n\geq2$. Under the corrected Statement the answer is yes, proved for every $n\geq2$ by the Ramsey corollary of the #548 theorem, with $R(T)\leq2n-3$ for odd $n$ (Progress, below). The problem's standing judges the corrected Statement. The $n=1$ refutations answer the site's wording (every $n$, the one-vertex tree included), not the corrected Statement (trees on $n\geq2$ vertices), so they do not count toward the problem's standing; they are credited here and on Alexeev's rejected claim page (2026).