How to Evaluate AI Stocks Like a Value Investor (Not a Speculator)

 

 

Every major technological revolution follows a familiar pattern. A genuinely transformative technology appears, investors get excited, money floods into the space faster than anyone can properly evaluate it, and prices detach from the underlying businesses. Eventually, the market sorts out which companies were real and which were just riding a narrative — and that sorting process is rarely gentle.

We saw it with railroads in the 1800s. We saw it with radio in the 1920s. We saw it again with the internet at the turn of the millennium. And right now, we're watching it happen with artificial intelligence.

If you're an investor trying to figure out which AI companies are worth owning and which ones are simply well-marketed stories, you're not alone. Most people buying "AI stocks" today are doing so with the same tools they'd use to evaluate any other company — and that's part of the problem. AI businesses come with financial and competitive dynamics that older valuation frameworks weren't built to handle.

Why Traditional Analysis Falls Short With AI Companies

Value investing, in the tradition of Benjamin Graham and Warren Buffett, has endured for nearly a century because its core principles don't go out of style: buy businesses, not tickers; understand what you actually own; demand a margin of safety; and let cash flow, not hype, guide your decisions.

The challenge is that applying those principles to AI companies isn't as straightforward as running a standard valuation model. A few things make AI businesses genuinely different to analyze:

R&D and compute spending distort earnings. Many AI companies are investing heavily in infrastructure and research well before that spending shows up as revenue. Traditional "owner earnings" calculations weren't designed with this kind of front-loaded capital intensity in mind, which means investors need to adjust how they read the numbers rather than take reported profitability at face value.

Moats look different now. For most of the 20th century, competitive advantage came down to brand recognition, distribution networks, or economies of scale. In AI, the moat conversation has expanded to include data access, talent concentration, switching costs embedded in workflows, and — increasingly — who actually controls compute capacity. A company can look strong on a balance sheet while quietly losing the more important competitive battle.

Winner-take-most dynamics are common. Some AI markets are shaping up to reward a small number of dominant players rather than supporting a wide field of competitors. That changes how you think about position sizing and risk, because being early in the "wrong" AI company can be just as costly as avoiding AI altogether.

Hype is louder than usual. Even experienced investors can struggle to separate a genuine multi-year competitive advantage from a temporary narrative advantage when the entire market is excited about the same theme at the same time.

None of this means value investing doesn't work for AI stocks. It means the framework needs an update — not a replacement.

Three Skills Every AI Investor Needs

If you strip away the noise, evaluating an AI company well comes down to three core competencies.

1. Financial literacy adapted for AI-specific accounting. Reading a standard income statement is one skill. Reading a 10-K for a company that's burning cash on compute infrastructure today while projecting profitability years out is a different and more demanding skill. You need to know which numbers to trust, which to adjust, and which to treat with healthy skepticism.

2. A framework for AI-era moats. Brand and distribution still matter, but they're no longer the whole story. Investors now need a systematic way to weigh data advantages, compute access, and talent retention alongside the traditional competitive factors.

3. Psychological discipline. This might be the hardest part. Understanding how markets tend to misprice transformative technology — both by overvaluing hype and by undervaluing genuine long-term winners — is often what separates investors who build real wealth from those who end up providing exit liquidity for someone else's story.

Where to Learn This Properly

Most of what's published about "AI investing" right now falls into two categories: breathless hype pieces predicting the next big winner, or newsletters selling stock picks with no real explanation of the underlying logic. Neither actually teaches you how to think.

That gap is exactly what led me to The AI Value Investor by Berg Codex, a 254-page, 34-chapter guide that applies Graham and Buffett's timeless principles to the specific challenge of evaluating AI companies today. It's not a stock-picking newsletter or a list of tickers — it's a structured investing education built around a few core ideas:

• How markets have historically mispriced transformative technology, and the psychology behind it
• Valuing AI companies using DCF, owner earnings, and free cash flow yield, adjusted for AI's accounting quirks
• An original seven-pillar framework for identifying genuinely durable AI companies across infrastructure, software, and other sectors
• The new competitive moats of the AI era, and the specific warning signs that tend to predict which AI companies fail
• Portfolio construction guidance covering position sizing, timing, and managing macro risk
• Fully worked case studies covering an AI chip company, an AI SaaS company, and an AI healthcare company

It also includes a full appendix of practical tools: a 50-point AI company evaluation checklist, a guide for building your own intrinsic value calculator, a due diligence worksheet, and a complete glossary of AI investing terminology. These are the kinds of resources meant to be used repeatedly, not read once and forgotten.

The Bottom Line

AI is a genuinely transformative technology, and that's exactly why it's dangerous to invest in carelessly. The companies that will matter in ten years are being built right now, alongside plenty of others that will quietly disappear. Telling the difference requires more than enthusiasm — it requires a framework.

If you want to build that framework rather than guess your way through the next decade of AI investing, The AI Value Investor is worth a look. You can find it here: thebergcodex.shop/products/the-ai-value-investor

This post is for educational purposes only and does not constitute financial advice. The author of the referenced book is not a licensed financial advisor. Always do your own research before making investment decisions.