Just caught something interesting about where the smart money is looking right now. Jim Cramer's been talking up two trillion-dollar plays in AI, and honestly, Wall Street seems to be on the same page with him.

So here's the thing - Jim Cramer recommended getting into Alphabet around $344 and Amazon near $239. Both have pulled back since then, but the analyst consensus is pretty clear: these are still undervalued at current levels. Alphabet's sitting at a median target of $385 from 74 analysts, suggesting 29% upside. Amazon's got 72 analysts calling for $285, which implies another 31% of room to run.

What's interesting is why Jim Cramer and the Street are so bullish. With Alphabet, it's the cloud computing story. Google Cloud revenue has accelerated three quarters straight, and their custom Tensor Processing Units are finally being monetized externally. Meta and Anthropic are already renting TPU capacity - that's real revenue starting to flow. Meanwhile, Google Search adapted with AI Mode and AI Overviews built on Gemini, and management says it's driving higher usage.

Amazon's angle is slightly different but equally compelling. AWS dominates with 41% market share, and CEO Andy Jassy is monetizing AI infrastructure as fast as they can build it. Cloud revenue just hit 24% growth - fastest in 13 quarters. Plus, their custom chips (Trainium and Inferentia) are now doing $10 billion annual run rate and growing triple digits. OpenAI just locked in a $138 billion deal for Trainium capacity.

The valuation question is where Jim Cramer's call gets interesting. Alphabet trades at 28x earnings, which sounds pricey until you remember they beat estimates by 15% average over the last six quarters. Amazon's at 30x with a 19% average beat rate. If that pattern holds, current prices look reasonable for patient investors.

Wall Street's expecting Alphabet earnings up 11% annually through 2027, and Amazon at 15% annually. Morgan Stanley even called Amazon the most underappreciated AI winner in their coverage. That's the kind of validation that gets attention.

The capital spending concern is real though - Amazon's planning $200 billion in capex for 2026, which spooked some people. But that's the cost of building the infrastructure these AI workloads need. Jassy's point is that AWS is literally running at full capacity monetizing what they deploy.

If you're thinking about this space, Jim Cramer's thesis basically comes down to: these companies have competitive moats, real AI revenue streams forming, and analyst estimates keep getting beaten. Not bad for trillion-dollar companies that supposedly have room to run.