The Magnificent Seven are trading at their cheapest valuation relative to the S&P 500 since 2015, according to Business Insider, citing Marta Norton, Chief Investment Strategist at Empower Investments.
Norton argues the pullback, which has seen the group fall as much as 19% from their October 2025 peak, has opened up a rare entry window for investors who believe in long-term AI adoption. “You are essentially paying the same type of valuation for these names as you would be just to get a broad collection of the US stock market,” Norton told Business Insider.
The timing of the selloff matters. The Nasdaq Composite dropped 7.1% in the first quarter of 2026, with all seven mega-cap tech stocks declining. Shares of all but two Mag 7 stocks were in the red year-to-date as of late February, as investors questioned whether the $600 billion-plus the hyperscalers plan to spend on AI infrastructure in 2026 can generate returns fast enough to justify the tab.
AI Stocks Q1 2026: A Quarter That Tested Conviction
The pressure on AI-linked equities has been building for months. Nvidia, which sits at the center of every hyperscaler buildout plans, saw its stock come under scrutiny after its latest earnings as investors asked whether AI capex spending might be approaching a ceiling. Microsoft, Alphabet, and Amazon have each signaled they have no intention of slowing their data center investment despite the short-term hit to margins.
The valuation compression has been dramatic in historical context. The Bloomberg Magnificent Seven Index, which tracks the seven mega-cap tech leaders, has not been this cheap relative to the broader market since 2015. For Norton, that historical parallel is significant: 2015 was also a period of skepticism about big tech ability to monetize the mobile internet transition, a story that proved wildly wrong over the following decade.
The AI infrastructure trade has parallels to early-stage buildout cycles in blockchain networks: high upfront capex, uncertain timelines to monetization, and infrastructure providers capturing value before application-layer companies do. Tokenized securities platforms — like the NYSE-Securitize infrastructure Frontierbeat reported on — are emerging as an alternative route to AI equity exposure.
AI Infrastructure Capex: The 600 Billion Dollar Bet
The four largest hyperscalers — Microsoft, Amazon, Alphabet, and Meta — are on track to spend a combined $650 billion to $690 billion on capital expenditures in 2026 alone. Meta alone has guided for $115 billion to $135 billion in AI infrastructure capex this year. Global IT spending overall is expected to surpass $6 trillion for the first time in 2026.
Norton framing is that the hyperscalers play both sides of the AI trade simultaneously: they are the largest buyers of AI infrastructure and its primary monetizers. That dual exposure gives them an advantage over pure-play chip suppliers, because they can spread the cost of the buildout across their own cloud services. Nvidia, which is the closest thing to a pure-play exposure in the group, has seen its revenue surge from $27 billion in 2022 to $216 billion in 2025.
Competitive dynamics in the AI space are shifting faster than traditional equity frameworks account for. The DeepSeek V4 launch earlier in 2026, which forced Google and Microsoft to accelerate their AI roadmaps amid the US-China technology race, is a reminder of that volatility — as Frontierbeat covered in its analysis of the situation.
Norton Picks: Nvidia, Amazon, Microsoft, Alphabet, Meta
Norton five highest-conviction AI names are Nvidia, Amazon, Microsoft, Alphabet, and Meta — explicitly excluding Apple and Tesla on the grounds that their AI monetization pathways are less direct. Her core argument is that AI infrastructure spending is not a bubble in the traditional sense because it is backed by real corporate balance sheets with genuine revenue streams.
The strategist acknowledges there is near-term downside risk. If interest rates remain elevated or if AI application revenue disappoints consensus expectations in 2026, the selloff could extend further. But she frames that scenario as an argument for buying incrementally rather than waiting for perfect clarity.
The broader macro context adds a final layer of uncertainty. Geopolitical tensions — particularly around semiconductor supply chains and US-China technology competition — have introduced a risk premium into AI equity valuations that did not exist two years ago. How that risk premium resolves will determine whether the Mag 7 re-rate back to historical averages or whether the current valuation window closes before the next earnings cycle confirms the AI revenue thesis.