Using AI to Pick ETFs: A Real Investing Use Case

After writing about “How Gen AI Can Help You Pick Stocks & Where It Falls Short”, I wanted to test it in the real world. No theory, practical use case; just me, AI, and the two most talked about ETFs in the moving sectors: the VanEck Semiconductor ETF (SMH) and BlackRock’s iShares Bitcoin Trust (IBIT).

The goal? See if AI could help me research, analyze, and confidently make an investment decision. The answer: absolutely yes, if used correctly. I didn’t just ask for a summary; I asked for a detailed analysis. I asked the questions I would usually research manually or leave for later due to time limitations. The result? Faster insights, a more innovative structure, and most importantly, confidence.

Why Even Use AI for Investing?

AI isn’t magic. But when used correctly, it becomes a high-powered research assistant that processes data faster than any human could. Whether you’re exploring sector trends, looking up ETF performance, or trying to decode what the online investing community is buzzing about, AI helps save time and gives you structure.

For me, it helped consolidate performance statistics, risk factors, and macroeconomic links (such as how AI is driving chip demand), as well as summarize conversations from platforms like X, primarily.

Of course, AI isn’t perfect. It can miss breaking news or give outdated information if you don’t frame your question well. But when used correctly, it can cut your research time from hours to minutes and give you a starting point that’s sharper than Google alone.

My AI-Powered ETF Research

Step 1: Asking Better Questions

Like most users, I started with a fundamental question: “Tell me about SMH and IBIT.” The AI provided me with general details on what they track, their fees (SMH: 0.35%, IBIT: 0.25%), and their top holdings. It was helpful, but surface-level.

To achieve better and more precise output, I became more specific in my questions. And continued refining my prompts.

That made all the difference.

The AI highlighted SMH’s potential growth from the surging demand for AI chips and flagged IBIT’s sensitivity to Bitcoin price movements and regulatory chatter. The quality of insights dramatically improved once I gave the AI context and focus.

Learning? AI is only as good as the question you ask.

Better prompts = Better Output.

Step 2: What I Learned About SMH

SMH’s core investment is in top semiconductor companies, such as NVIDIA, TSMC, and ASML, the very companies powering the AI revolution. The AI Assistants showed me that SMH has delivered nearly 30% annualized returns over the past five years and has been regaining momentum for over three months, driven by insatiable demand for chips across AI, gaming, EVs, and data centers. It also reminded me that SMH is concentrated, meaning if something hits NVIDIA or the chip cycle turns, the ETF could drop fast.

AI tools summarized social chatter from X, revealing a mix of bullishness and warnings about stretched valuations. That nuance helped me to calibrate my expectations.

Bottom line: SMH looks strong for long-term investors who can handle some volatility. I added it to my watchlist and eventually included it in my portfolio.

Step 3: What I Learned About IBIT

IBIT offers direct exposure to Bitcoin, eliminating the need for a wallet or exchange. With Bitcoin up more than 50% this year and significant funds joining the party, it’s easy to see why IBIT got my attention.

AI Assistants helped compare IBIT with other Bitcoin ETFs, noting its low 0.25% fee and strong tracking accuracy. It also highlighted key risks: Bitcoin’s extreme volatility, regulatory overhangs, and sentiment swings that can drive price moves overnight.

On X, investor sentiment was split. Some viewed Bitcoins as digital gold in the AI age. Others warned of FOMO buying at the top.

For me, the takeaway was clear: IBIT is a high-risk, high-reward play, best suited for risk-tolerant investors with a multi-year view.

Step 4: Blending AI With Human Thinking

AI assistants gave me clarity, structure, and speed. However, it didn’t influence my investment decision.

That was the most important lesson. AI is not a fortune-teller; it’s a lens. It showed me trends, summarized chatter, and surfaced valid data points. But I still had to align the insights with my personal goals, time horizon, and risk tolerance.

AI Assistants helped me slow down and think more clearly, which ironically made me a better human investor.

Want to Use AI for Your Own ETF Research? Start Simple

You don’t need fancy tools or coding skills. You can start right now. Here’s how: You don’t need to be an engineer or have a paid subscription to Bloomberg. Here’s how to start:

1. Pick a free AI tool: ChatGPT, Grok, Gemini, or Perplexity. ( I have used all of them at different times for different analyses)
2. Start simple: Ask, “What is [ETF] and what are its top holdings?”
3. Get specific: Ask, “What is the 1–3-year outlook, risks, and recent performance?”
4. Layer in social: Ask, “What’s the sentiment on X about [ETF]?”
5. Double-check everything: AI gives you direction. Verify facts using ETF websites like VanEck (for SMH) or iShares (for IBIT).

Final Thoughts: Man + Machine = Smarter Investing

Using AI to analyze SMH and IBIT was more than just a neat experiment — it was a genuine investment process that saved time, reduced guesswork, and helped me stay focused.

SMH let me ride the AI chip wave. IBIT lets me explore crypto exposure with guardrails. In both cases, AI was my co-pilot, not the pilot.

And that’s the future of investing, I believe in. Not blindly following algorithms, but using machine intelligence to make sharper, more confident decisions.

Try it. Ask better questions. Let AI help you invest smarter.

Disclaimer: Investing involves risks. Past performance is not indicative of future results. Please do your research or consult a financial advisor. Data as of July 2025.