Nvidia is executing a multi-front expansion — RTX Spark into PCs, Rubin platform into next-gen data centers, physical AI open-source tools, and Jensen Huang personally diplomating with Korean tech giants — while GOOGL sits at a $380.34 price with +125.79% over 52 weeks, the single best performer in the Mag 7 over that stretch. NVDA is down -1.45% today at $211.14 despite the news barrage, suggesting the market is digesting geopolitical chip export headwinds alongside the growth narrative. Both names deserve a close look right now — but for very different reasons.
Let's start with Nvidia, because Jensen Huang is currently running one of the most aggressive multi-theater expansions any CEO has attempted in recent memory. This week alone: RTX Spark launches as a direct assault on Intel and AMD's PC chip turf, targeting AI agents on personal devices using Arm-licensed architecture. Jensen is flying to South Korea to meet LG and Samsung leadership to explore AI and robotics partnerships. NVDA and TSMC announced AI-in-fabs collaboration. A major open-source physical AI toolkit dropped. Nebius got the Jensen halo effect at Computex. This isn't a company resting on data center dominance — it's a platform company making simultaneous moves across PC, robotics, enterprise storage certification, and international supply chain diplomacy. The Rubin platform is the next revenue catalyst to watch carefully.
And yet NVDA closed at $211.14, down -1.45% today, on 282.7 million shares. The headwind is real and it's called export controls. The DOC is closing a loophole that allegedly allowed Blackwell and Rubin processors to reach Chinese entities through subsidiaries outside mainland China for nearly a year. This matters because China remains a massive addressable market that Nvidia is increasingly locked out of. Every regulatory tightening incrementally shrinks the TAM ceiling and creates execution risk on forward estimates. AMD's MI350 chips are caught in the same net, so this isn't a competitive disadvantage per se — but it does constrain the entire sector's China exposure narrative.
On valuation: NVDA trades at 32.38x TTM P/E on $253.5B in trailing revenue with a $5.11T market cap, 20.17x P/S, and a beta of 2.244 — the highest in the Mag 7 by a wide margin. That beta tells you exactly what the risk profile looks like heading into a macro environment the Fed is still characterizing as historically expensive. The profit margin of 62.97% is exceptional and the operating leverage story is intact, but at these multiples, you need the Rubin ramp to deliver, PC penetration to accelerate, and no further export tightening. Three simultaneous execution requirements is a crowded ask.
Now shift to Alphabet. GOOGL at $380.34 is up +20.77% YTD and +125.79% over 52 weeks — that 52-week figure is not a typo. It is the best-performing Mag 7 name on a trailing year basis, and it's doing it at a 28.99x TTM P/E on $422.5B in revenue with a $4.61T market cap. Compare that to NVDA's 32.38x P/E on $253.5B revenue — GOOGL is generating significantly more top-line at a cheaper multiple. The EV/EBITDA of 20.80x is reasonable for a company at this scale with its AI infrastructure deployment. The 37.92% profit margin is a machine. GOOGL is also issuing debt strategically to fund AI capex — a sign of disciplined capital allocation, not distress. The Enlight renewable energy deal reinforces that Google is thinking 10-15 years out on infrastructure cost management.
The one genuine overhang on GOOGL today is the DOJ insider trading case involving an employee allegedly using internal data on Polymarket. This is a governance and reputational nuisance, not a fundamental impairment — but these stories have a way of generating regulatory scrutiny at exactly the wrong moment, particularly given that Alphabet is already navigating antitrust pressure on its search monopoly. The market sold GOOGL -2.51% today, which feels like a combination of this headline noise and broader risk-off sentiment tied to Fed inflation signals. I'd characterize that dip as noise over signal for long-term holders. The core search + cloud + AI flywheel remains intact.
Putting it together: I'm running MIXED-to-BULLISH on this pair with asymmetric conviction levels. GOOGL is the cleaner long — best 52-week performer in Mag 7, reasonable multiple for its earnings power, diversified revenue, and an AI infrastructure story that's already monetizing. NVDA is the higher-conviction story IF Rubin delivers and export controls don't escalate, but with beta of 2.244 and a crowded catalyst queue, sizing discipline matters. Both are worth owning; GOOGL is the more comfortable hold heading into the second half of 2026.