[Archive] Giving ChatGPT Full Control Over a $100 Micro-Cap Portfolio
Can AI turn $100 into alpha with complete freedom? Let's find out.
Note: this experiment is no longer active. I recently had a better idea to put ChatGPT and DeepSeek against eachother with the same concept. Check it out on my profile!
Welcome to my real-money experiment: Can an LLM beat the market with just $100?
For the next 6 months, I’m testing whether a large language model like ChatGPT (4o to be specific) can generate real profit using only U.S.-listed micro-cap stocks. I’ve set strict rules:
Only full-share positions
Market cap under $300M
$100 total budget
6-month time frame (6/23/25 – 12/23/25)
I’m giving the AI complete control over what to buy, how much, when to sell, and how to manage risk. This isn’t a paper-trading experiment — I’m using real money so I stay invested and accountable. I will only reach out to provide data and get decisions, nothing else.
I’ll post regular updates with price movements and ask the AI if any changes are needed.
Here's the original prompt I gave:
“You are a professional-grade portfolio strategist. I have exactly $100 and I want you to build the strongest possible stock portfolio using only full-share positions in U.S.-listed micro-cap stocks (market cap under $300M). Your objective is to generate maximum return from today (6-23-25) to 6 months from now (12-23-25). This is your timeframe, you may not make any decisions after the end date. Under these constraints, whether via short-term catalysts or long-term holds is your call. I will update you daily on where each stock is at and ask if you would like to change anything. You have full control over position sizing, risk management, stop-loss placement, and order types. You may concentrate or diversify at will. Your decisions must be based on deep, verifiable research that you believe will be positive for the account. Now, use deep research and create your portfolio.”
And here’s what the AI came up with after about 10 minutes of deep research:
The full report is pretty long, (Full Reply: Response), but here’s a summary.
Portfolio:
MYO – Myomo Inc.
14 shares @ $3.00 → Limit Buy | Total $42
Medtech company with surging sales from Medicare expansion. Q2/Q3 earnings could confirm breakout growth.
CADL – Candel Therapeutics
5 shares @ $5.60 → Limit Buy | Total $28
Biotech with positive Phase 3 cancer data. Major survival results due Q4 — high upside if results hold.
CPSH – CPS Technologies
10 shares @ $3.00 → Limit Buy | Total $30
Industrial turnaround play. Record revenue, new defense contracts, and real profitability.
So far I haven’t bought anything, but I have put the orders in.
My initial thoughts are pretty positive. While I think the AI won’t be able to find truly hidden undervalued stocks, I think it will make pretty thought-out decisions about what’s already online. My estimate is that the total value will grow to $105 - $110 (5% - 10%). Considering the chosen assets are BioTech, much bigger swings either way are likely.
Stay tuned! Every day, I’ll update the prices and let the AI react: rebalancing, taking profit, or holding.
Let’s see if AI can be a hedge fund manager.