Erdős Problem #320 — Let S(N)S(N) count the number of distinct sums of the form ∑n∈A1n\sum_{n\in A}\frac{1}{n} for A⊆{1,…,N}A\subseteq \{1,\ldots,N\}.

At least 52 years old · documented by

Let S(N)S(N) count the number of distinct sums of the form ∑n∈A1n\sum_{n\in A}\frac{1}{n} for A⊆{1,…,N}A\subseteq \{1,\ldots,N\}. Estimate S(N)S(N).

References

Progress summary

Refreshed
Claimed solved

A new AI-assisted proof claims to settle the estimate at the correct scale, but independent mathematical verification has not yet appeared.

Erdős Problem #320 asks how many different totals can be formed by choosing any collection of the first NN unit fractions. The established bounds had matching broad shape but did not settle the estimate precisely.

Known results

  • Bleicher and Erdős proved lower and upper bounds of iterated-logarithmic scale for log⁡S(N)\log S(N).
  • Bettin, Grenié, Molteni, and Sanna (2025) strengthened the lower bound to
log⁡S(N)≥Nlog⁡N(2log⁡2(1−3/2log⁡kN)∏i=3klog⁡iN).\log S(N)\geq \frac{N}{\log N}\left(2\log 2\left(1-\frac{3/2}{\log_kN}\right)\prod_{i=3}^{k}\log_iN\right).

August 2026 claimed proof

The Erdős Problems entry reports that GPT 5.6 Sol, prompted by Young, Zhu, and Luo, obtained an upper bound matching the known lower-bound order. A related repository claims a more precise asymptotic and a Lean 4 formalization, but its four axioms encode imported results and computational certificates; the claimed resolution remains unverified.

Current status (as of August 2026): The classical order-of-magnitude bounds are established, while the claimed matching proof and sharper asymptotic remain unconfirmed.

  • GPT-5.6 SolOpenAIsolved2026-07-01evidence
Sources

Solutions 0

No solutions have been posted yet.