AI Stocks: Is It Too Late to Get In?

My Coworker Made 3 Million Won on NVIDIA

At lunch, a colleague casually mentioned, "I bought NVIDIA last year -- up 47% now." Lost my appetite. (I'd considered buying it last year too. Didn't, because "hasn't it already run up too much?")

That evening, I installed a stock trading app. Had an account already but hadn't opened it in a year and a half. Balance: 12,340 won (~$9). Idle cash that had been sitting there.

What Even Are "AI Stocks"?

Did some research. AI-related stocks fall into roughly three buckets.

Semiconductors. NVIDIA, AMD, TSMC. Companies making the GPUs that train AI. The most direct play.

Cloud/Infrastructure. Microsoft, Google, Amazon. AI needs servers, and these companies provide them. Indirect but solid beneficiaries.

Software. Palantir, C3.ai, and Korean companies like SelvasAI and Saltlux. Companies building actual AI services. Highest risk bucket.

Starting with 487,000 Won

I didn't have the courage for big money. Took 500,000 won from my salary's discretionary budget (ended up being 487,000 won after transfer fees and rounding -- the details that never work out cleanly).

Bought 1 share of NVIDIA (around $127 at the time), 0.3 shares of Microsoft (fractional shares), and 150,000 won in a Korean AI ETF. Calling it "diversified" at this amount is generous, but at least it's spread across three things.

Month One: Checking Addiction

I opened the app maybe 8 times a day. Waking up, on the subway, at lunch, in the bathroom, heading home... Stock goes up 0.5% and I feel good. Down 0.3% and anxiety hits.

First month's gains: +11,200 won (+2.3%). Technically a profit, but 11,200 won is one lunch. If it had been -11,200 won, losing one lunch, the emotional response would've been completely different. Human psychology is weird.

The Correction Hit

Month two, "AI bubble" discourse emerged and tech stocks broadly declined. NVIDIA dropped 5% in a single day. My portfolio flipped to -3.7%. In dollar terms, about -$13, but emotionally it felt like -$130.

"Should I sell now?" Lost sleep over this thought. Over $350. If this is how stressed I get over $350, big money would've been genuinely dangerous. Didn't sell in the end. "It's a long-term investment." (Whether that's genuine long-term thinking or just inability to cut losses, still unclear.)

Does Being a Developer Help?

People sometimes say "as a developer, you understand AI better so you have an investing edge." Half true, half not.

True part: I naturally consume tech news, so I have intuition about which technology is genuinely useful versus marketing. "My own company is evaluating this tech, so real demand exists" -- that kind of feel.

False part: understanding technology and predicting stock prices are entirely different domains. Great tech companies can still see falling stock prices. The "great tech" is already priced in.

Three Months In

Current return: +4.8%. In dollars, about +$17. Three months to earn $17. If I calculate the hourly rate factoring in time spent checking prices... better not to calculate. The hours spent on price-checking alone are substantial.

Did I Learn Anything?

Yes.

First, starting small was genuinely the right call. $350 meant I could lose sleep but not lose my mind. This was tuition for learning investment psychology. Cheap tuition.

Second, on whether AI stocks are "too late": I don't know. Really don't know. People said "too late" a year ago. They're saying it now. We'll know in five years.

Third, stocks are stocks and work is work. Checking prices while coding cannot become a habit. I turned off app notifications.

I'm debating whether to add another $350 next month. But writing this post, I'm thinking "wouldn't that money be better spent on server costs for side projects?"