How We Pick AI Stocks: A Fundamentals and Momentum Method
There are well over a hundred listed companies with a credible claim to being an "AI stock." Most retail investors pick from headlines: a tip here, a Reddit thread there. They end up holding names they can't actually explain. There's no single right way to do this. This is the one we've settled on. It's the method AI Investing Pulse uses to cut a large, noisy universe down to a ranked shortlist, built on two signals: company fundamentals, and momentum (how a stock is performing against its AI peers). Fundamentals carry the most weight and momentum confirms, but both have to line up. You can apply the same five steps yourself.
Step 1: Define the universe across the AI value chain
Don't treat "AI stocks" as one blob. Map every candidate to where it sits in the AI value chain. We track more than 200 AI-exposed companies across five layers:
Layer | What it is | Examples of the type |
|---|---|---|
Build | Hardware, semiconductors, data centres | chipmakers, fabrication equipment |
Connect | Networks, cloud, data integration | hyperscalers, networking |
Develop | AI dev tools, security, compliance | MLOps, observability, security |
Deploy | Applications, SaaS, end-user AI | enterprise and consumer AI software |
Power | Energy, cooling, quantum | data-centre power, cooling, next-gen compute |
Mapping by layer does two things: it shows you what you're actually betting on, and it stops you over-concentrating in one part of the chain (most retail AI portfolios are quietly 80% "Build").
Step 2: Score the fundamentals (this carries the most weight)
For each company, build a single Fundamental Score out of 100 from three inputs, weighted deliberately toward growth:
Component | Weight | What it measures |
|---|---|---|
Growth | 65% | Is revenue accelerating, and is AI a real driver of it? |
Financial Quality | 25% | Margins, cash flow, balance-sheet strength |
Valuation | 10% | Does the price make sense given the growth? |
Growth dominates because in AI the winners are defined by durable demand, not by being cheap. Valuation still matters. It's the brake, but it doesn't lead.
Step 3: Check momentum (relative strength vs the AI sector, not the S&P)
A stock can be fundamentally strong and still be ignored by the market. Relative Strength (RS), scored out of 100, is our momentum measure. It captures how a stock is performing against the other AI stocks, not against the whole market. RS of 90 means it's outperforming most of the AI universe; 20 means it's lagging. Measuring RS within the AI sector is the point: it tells you whether the market is currently rewarding this AI name versus its real peers.
Step 4: Combine into one number
Add the two together for a single AIIP Score out of 200 (Fundamental /100 + RS /100). A high combined score means a strong business that the market is also rewarding: agreement on both dimensions. That's the number to rank by.
Step 5: Filter to a shortlist
The method narrows the whole universe to a shortlist rather than working from all of it. What's left is the names that clear both bars: strong fundamentals and relative strength in the AI sector's top tier (RS at or above 80). That intersection, fundamentally sound and market-confirmed, is a short, workable watchlist, not a list of a hundred tickers.
FAQ
What's the best way for a retail investor to pick AI stocks? There's no single best way. Here's the approach we use: we map candidates across the AI value chain, score each on fundamentals (weighted toward growth), then check relative strength against the AI sector. We rank by the combined score, and the shortlist narrows to the names that are strong on both fundamentals and market performance. You can apply the same steps yourself.
Fundamentals or momentum: which matters more? In our method, both matter, in that order. Fundamentals (growth, financial quality, valuation) carry most of the weight, around two-thirds on growth alone. Relative strength acts as a confirmation layer: it shows whether the market currently agrees, rather than leading the decision.
How many AI stocks should I actually track? A universe of 200+ is enough to cover the AI value chain without missing the second-tier names. The method then works from a filtered shortlist: the names that clear both the fundamental and relative-strength bars, usually a few dozen at most.
Why score relative strength against AI stocks instead of the whole market? Because in a sector running as hard as AI, beating the S&P is a low bar. What matters is whether a stock is leading or lagging its actual peers: the other AI names competing for the same capital.
Disclaimer. AI Investing Pulse provides research, data, and educational frameworks for informational purposes only. It does not constitute investment advice, a personal recommendation, or an invitation or inducement to engage in any investment activity. AI Investing Pulse is not authorised or regulated by the Financial Conduct Authority (UK), is not registered with the U.S. Securities and Exchange Commission or any U.S. state regulator, and is not registered with any Canadian provincial securities commission or CIRO. Nothing here is tailored to your individual circumstances. The value of investments can fall as well as rise and you may get back less than you invested; past performance is not a guide to future results. Always consult a qualified, regulated financial professional before making investment decisions. Our scoring methodology is rules-based and explained in full on this page.
