# Claude aids stock analysis but cannot replicate 21.8% annual returns, investigation finds

Digest AI · Marketing & Small Business · published 2026-10-27T07:00:00Z

Canonical: https://digestai.news/story/claude-aids-stock-analysis-but-cannot-replicate-21-8-annual-returns

## Summary

An X post claimed a former entrepreneur used **Claude** to replicate an investment strategy yielding **21.8% annual returns** over **12 years**, beating the S&P 500 in **11 of 12 years** with just **2.9 trades per year**. The post suggested beginners could adopt this method by having Claude research company documents, create investment memos, and wait for favorable prices. However, a verification found **no verified basis** for the 21.8% figure, as the investor’s track record came from an unverified podcast (*My First Million*) and lacked third-party audits. The **2.9 trades/year claim** also conflicted with the podcast’s data, which showed **22 trades over ~7.5 years**—not 12 years—averaging **2.93 trades/year** instead of 1.83 if stretched over 12 years.

Claude’s role in the process was limited to **organizing corporate filings** (e.g., summarizing **securities reports** from Japan’s EDINET) and **creating structured memos** for investment decisions. While Claude can extract data from PDFs and flag inconsistencies, it **cannot guarantee accuracy**—misinterpreted units (e.g., 12,500 million yen vs. 12.5 billion yen) or misaligned accounting terms (e.g., operating profit vs. net income) risk errors. The post’s method of **setting price thresholds** for purchases also proved risky, as external shocks (e.g., scandals) could invalidate prior analysis. The investigation concluded Claude **enhances due diligence** but **does not replace human judgment** or replicate elite performance. The author noted stock investing differs from AI side hustles (e.g., content creation) due to **principal risk** and **unrealized gains**, making it unsuitable for stable monthly income goals like the series’ **¥10,000/month target**.

## Key points

- Claude can summarize corporate filings and draft investment memos but lacks verified accuracy for predicting returns
- Unverified podcast data underpins 21.8% annual return claim; trades/year figures conflict with original sources
- Human oversight is critical: AI-extracted data may misinterpret financial terms or miss external risks

## Why it matters

This debunks hype around AI-driven investing, clarifying Claude’s role as an **analytical aid**, not a profit guarantee. For retail investors, it highlights risks of over-reliance on AI for high-stakes decisions without human verification.

## Sources

1. [AI Side Hustle Verification #44 Can you beat the S&P 500 with Claude? Verifying the investment strategy and AI analysis capability behind '21.8% annual return, 11 wins in 12](https://note.com/tsuwami/n/n6aaeee1433b9?hl=en) (note.com, 2026-10-27)

## Cite

Digest AI, "Claude aids stock analysis but cannot replicate 21.8% annual returns, investigation finds", 27 October 2026, https://digestai.news/story/claude-aids-stock-analysis-but-cannot-replicate-21-8-annual-returns

---

Written by Digest AI's editorial model from the linked sources; the sources are the record. Headlines, digests and key points are written by Digest AI and may be quoted with a link to the story page. Linked articles belong to their publishers. Terms: https://digestai.news/terms#reuse
JSON: https://digestai.news/story/claude-aids-stock-analysis-but-cannot-replicate-21-8-annual-returns.json