People Use AI
Finance

Competitive moat analysis using a structured framework

Use an LLM to analyze a company’s competitive moat using a predefined framework (e.g., Morningstar moat sources like brand, switching costs, network effects, cost advantages, intangible assets).

Why the human is still essential here

The investor must supply accurate context and evidence, and make the final qualitative judgment about moat strength rather than relying on generic model narratives.

How people use this

Morningstar-style moat scorecard draft

Feed the model your notes and key quotes and have it populate a moat scorecard across brand, switching costs, network effects, cost advantage, and intangibles with a short evidence summary per category.

ChatGPT

10-K evidence mapping to moat drivers

Paste relevant 10-K sections (competition, segments, customers, IP) and ask the model to map specific excerpts to each moat driver and flag where evidence is weak or missing.

Claude

Competitor landscape and differentiation summary

Use a research-focused LLM to compile a competitor list and summarize differentiation claims that relate to moat sources (e.g., platform effects, distribution, cost structure) with citations.

Perplexity

Community stories (1)

Reddit

I've been using ChatGPT for stock research during lunch breaks!

I started using ChatGPT for stock research about six months ago, mostly out of boredom between meetings. After a lot of trial and error, I've narrowed it down to a handful of prompts that consistently give useful output.

The most useful ones:

- Devil's advocate prompt — Tell ChatGPT to argue against a stock you're excited about. It's surprisingly good at poking holes when you explicitly ask for the bear case. Helped me avoid a couple of impulsive buys.

- Earnings report summarizer — Paste the key numbers from a quarterly report and ask it to explain them in plain English + flag what stands out. Saves a ton of time vs reading the full transcript.

- Competitive moat analysis — Using the 5 moat types (brand, switching costs, network effects, cost advantages, intangibles). Forces a structured evaluation instead of vibes.

The biggest lesson: ChatGPT is great for understanding stocks, terrible for predicting them (AI is not yet READY for that and I am not sure if it's ever going to be). It'll confidently give you wrong numbers too, so you absolutely have to verify everything against Yahoo Finance or SEC filings.

I wrote up all 10 prompts with examples and tips here if anyone wants the full list:

https://www.boredom-at-work.com/chatgpt-stock-research/

What prompts do you use ChatGPT for when it comes to finance or investing? Curious if anyone's found good ones I'm missing.

B
Bubbly_Ad_2071Retail investor
Mar 4, 2026