Michael Fritzell writes Asian Century Stocks, a research service covering undercovered equities across Asia. He spent sixteen years on the buy side in Hong Kong, Singapore, and Indonesia before leaving in 2021 to publish full-time, and now writes deep dives on companies that receive almost no analyst coverage.
This discussion took place live on September 2nd, 2026, on the MicroCapClub Community. Join MicroCapClub and unlock the ability to listen and participate live in these discussions - https://microcapclub.com/#join
In this conversation, Michael explains how he uses generative AI in his research process: the saved prompts he runs on any new ticker, including a bull-versus-bear project and one built to surface red flags in the accounts, and where he refuses to use these tools at all, namely calculations and writing. He also talks about where he finds ideas that AI can't commoditize, mainly insider transaction data and broken IPOs in Asia, and why he shifted from buying low P/E stocks to targeting a prospective IRR and selling once it falls below his cost of capital. He argues that faster research doesn't automatically mean better returns, which leads to the question of why an investor would cover two hundred stocks a year instead of twenty.
✉️ Share your feedback - david@microcapclub.com
✉️ David’s X (Twitter) - https://x.com/Valuehunte
Chapters
00:00 Intro
02:01 Is AI making us dumber?
02:56 Where you shouldn't outsource to AI
03:17 Benchmarking ChatGPT, Gemini, Claude and Grok
05:30 Getting up to speed on a company in hours
07:23 The bull vs bear prompt
08:19 What he won't use AI for
09:05 Why AI writing loses the reader
09:50 Brainstorming, obscure data and summaries
11:21 Using Claude for error checking
12:02 Writing by hand in an age of AI slop
13:26 What AI changed in his own business
15:44 From low P/E to prospective IRR
17:41 Knowing when to sell
18:18 Idea generation: insider transactions
21:04 Broken IPOs and borrowing from others
21:23 Red flags prompt for Asian due diligence
22:46 Why he still writes 40-page deep dives
23:23 How to prompt: give it context
24:41 Can AI run the whole investment process?
26:40 Why cover 200 stocks instead of 20?
Disclaimer: All content on this channel is for discussion, education, entertainment, and illustrative purposes only and SHOULD NOT be construed as professional financial advice, solicitation, or recommendation to buy or sell any securities, notwithstanding anything stated on this channel. There are risks associated with investing in securities. Loss of principal is possible. Past performance is not a predictor of future investment performance. Ian Cassel and the guests on this channel are not responsible for investment actions taken by viewers. Should you need such advice, consult a licensed financial advisor, legal advisor, or tax advisor. You agree to verify all information yourself before investing. Any past performance discussed during this program is no guarantee of future results. Investing involves risk and possible loss of principal capital; please seek advice from a licensed professional. All views expressed are personal opinions as of the date of recording and are subject to change without the responsibility to update views. No guarantee is given regarding the accuracy of the information on this channel. Releasees undertake no obligation to provide accurate or sound investment statements. You waive any and all duties that may exist flowing from you to any Releasee. You agree not to hold any Releasee liable for any possible claim for damages arising from any decision you make based on information or other content on the Channel.
