By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panelists agreed that AI is a significant growth opportunity but expressed concerns about over-optimistic valuations, regulatory risks, and potential cannibalization. They also debated the sustainability of Nvidia's hardware-software moat and the potential impact of custom silicon from hyperscalers.
Risk: Regulatory fragmentation and potential cannibalization of high-margin legacy software were the most frequently cited risks.
Opportunity: AI monetization across ads, software, and data-center hardware was seen as a significant opportunity for growth.
This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →
Finding core tech stocks to build a portfolio around can be a smart idea for investors. This sector has created the majority of value in the market for the past decade, and that will likely continue for the next decade as artificial intelligence (AI) innovations and breakthroughs occur.
Three that I think qualify for this segment are Alphabet (NASDAQ: GOOG) (NASDAQ: GOOGL), Microsoft (NASDAQ: MSFT), and Nvidia (NASDAQ: NVDA). Each of these looks like an excellent building block for a portfolio, and I think all will be a smart pick over the next decade.
Where to invest $1,000 right now? Our analyst team just revealed what they believe are the 10 best stocks to buy right now, when you join Stock Advisor. See the stocks »
Alphabet is first on my list for a good reason: It's the most solid of the three. Alphabet has a multi-pronged approach to AI, and its strategy so far has proven fairly solid.
First, it's integrating AI into its core Google Search product. This has made Google the go-to for quick AI-generated information on a topic, and it's a feature loved by the billions who use the product. Alphabet uses its own AI model to do this, leading to the second prong. Thanks to a strong generative AI model in Gemini, it can control its destiny with AI and its business.
Lastly, Alphabet has a thriving cloud computing wing, with Google Cloud growing faster than any of the major cloud providers in Q1. Google Cloud gives AI developers and companies a place to run AI workflows, so even if Alphabet's model doesn't come out on top, Google Cloud won't be a flop, either.
This multifaceted approach has so far proven successful for Alphabet, and its stock has doubled over the past year. Still, there is more upside ahead for Alphabet if it can maintain its current growth pace, and I think it will be an AI force to be reckoned with for years to come.
Microsoft's AI approach is very similar to Alphabet's, except that it isn't developing its own AI model. Instead, it has chosen to partner with OpenAI, the makers of ChatGPT. Microsoft owns about 27% of OpenAI, so it has a vested interest in its success. Microsoft's Azure cloud computing platform remains neutral and offers developers countless large language models to deploy and use. However, Microsoft has integrated OpenAI's products into all of its existing business productivity software via Copilot.
This has been a successful approach so far, but the stock hasn't responded to Microsoft's results as it has with Alphabet's. Microsoft is down over 30% from its all-time high, and it looks like a screaming deal as the future is bright.
Last is Nvidia, which may seem like an odd choice. Current market sentiment is that Nvidia's stock will decline once the AI build-out is wrapped up.
While that's a fair take, it ignores the fact that computing units in data centers have relatively short lifespans and need to be replaced every couple of years. Furthermore, Nvidia will likely keep innovating and developing new computing units with enhanced capabilities that can cut costs and improve performance, which could justify upgrading old systems.
Nvidia is interwoven into nearly every AI product, and it will remain a vital company in the industry long after the initial build-out is complete. However, the market isn't pricing any of that into Nvidia's stock.
The stock trades for a mere 22.3 times forward earnings, and less than 16 times next year's earnings. That's a major bargain that doesn't come around very often. With Nvidia being a core part of the AI build-out still expected to last for multiple years, the stock is a great buy now and a solid one to build a portfolio upon.
Before you buy stock in Alphabet, consider this:
The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Alphabet wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.
Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $382,359! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,201,390!
Now, it’s worth noting Stock Advisor’s total average return is 883% — a market-crushing outperformance compared to 205% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.
**Stock Advisor returns as of June 26, 2026. *
Keithen Drury has positions in Alphabet, Microsoft, and Nvidia. The Motley Fool has positions in and recommends Alphabet, Microsoft, and Nvidia. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.
Four leading AI models discuss this article
"Valuations price in a perpetual AI-driven growth arc; a softer AI adoption curve or a slower cloud cycle could lead to material re-rating."
Storyline: The piece positions Alphabet, Microsoft, and Nvidia as core AI bets, tying Gemini, Copilot, and hyperscale compute into a long-running growth thesis. Bullish takeaway: AI monetization across ads, software, and data-center hardware could compound for a decade, supporting 3–10% annual earnings growth plus multiple expansion as hardware/AI tooling becomes standard. But a blind spot is the risk that AI upside is over-optimistic and valuations bake in a perpetual AI-driven supercycle. If cloud demand slows, ad revenue softens, or regulation limits AI monetization, these stocks could re-rate. Nvidia’s cycle may peak; Alphabet/Microsoft face monetization challenges beyond marginal AI gains. Still, the setup is credible but not risk-free.
AI demand could disappoint, and valuations may compress if data-center capex slows or a macro downturn hits enterprise budgets. Nvidia’s cycle could peak, and Alphabet and Microsoft still face monetization and regulatory headwinds that could cap upside.
"The article uses inaccurate price data for Microsoft and ignores the high probability of multiple contraction if AI software monetization fails to meet aggressive enterprise expectations."
The article's premise relies on the 'AI infrastructure build-out' thesis, but it fundamentally mischaracterizes the valuation landscape. Claiming Microsoft is down 30% from all-time highs is factually incorrect; MSFT has been trading near record levels, suggesting the author is working from stale data. While Alphabet and Nvidia benefit from massive capital expenditure cycles, the real risk is 'AI ROI fatigue.' If enterprise software adoption of Copilot or Gemini doesn't translate into tangible margin expansion by 2025, we will see a significant contraction in forward P/E multiples. Investors are paying for a decade of growth today; any hiccup in cloud growth rates will lead to a painful re-rating.
The 'AI ROI fatigue' argument ignores that these firms are essentially utility-like monopolies; even if AI growth slows, their core businesses—Search, Windows, and Data Center hardware—remain indispensable cash cows.
"The article mistakes 'necessary for AI' with 'will outperform because of AI,' ignoring that consensus mega-cap holdings often underperform due to already-priced-in expectations and valuation compression risk."
This article conflates 'AI exposure' with 'AI winner,' a dangerous conflation. Yes, GOOG, MSFT, and NVDA are embedded in AI workflows—but that's precisely why they're priced for perfection. GOOG's search dominance faces real cannibalization risk from AI-native competitors; MSFT's Copilot monetization remains unproven at scale; NVDA's 22.3x forward P/E assumes continued capex spending that may plateau. The article ignores that 'core holdings' often underperform because consensus has already priced in the upside. The 10-year horizon is also a tell: it's long enough to paper over near-term valuation risk.
These three companies do control the AI infrastructure layer and have proven execution track records—NVDA's moat in chip design is real, MSFT's enterprise relationships are sticky, and GOOG's search flywheel remains formidable. If AI adoption accelerates faster than expected, all three could justify current multiples.
"Nvidia’s valuation assumes sustained pricing power that new competitors and capex normalization are likely to erode within three years."
The article frames Alphabet, Microsoft, and Nvidia as durable AI compounders at attractive prices, but it underplays structural risks. Alphabet faces ongoing antitrust scrutiny that could force search divestitures; Microsoft’s OpenAI economics remain opaque and its Copilot monetization is still early; Nvidia’s 22x forward earnings embeds assumptions of multi-year 40%+ growth that hyperscaler budget cycles and AMD/Intel catch-up could invalidate. A decade hold requires these moats to survive both regulatory and technological disruption that rarely spares even dominant tech names.
Even with slower growth, these companies’ scale in data centers and cloud could still generate returns well above the market if AI workloads keep expanding, making the ‘bargain’ label stick longer than skeptics expect.
"Regulatory fragmentation and data-localization constraints could slow AI monetization and cloud growth more than ROI fatigue, risking multiple compression."
Gemini's misstatement aside, the bigger flaw is underplaying regulatory fragmentation as a real limiter on AI monetization. If data localization, antitrust concessions, or export controls slow cross-border data flows, cloud demand and Copilot-like products could stall even with healthy capex. That risk could drive multiple compression even if AI workloads stay robust. In short: valuation risk isn't just ROI fatigue; policy friction could be a bigger headwind.
"Regulatory hurdles and software moats create a defensive barrier that protects incumbents from external competition, shifting the primary risk to internal product cannibalization."
Claude, you’re right on the 'priced for perfection' trap, but you’re missing the hardware-software feedback loop. Nvidia isn't just a chip designer; they are building the CUDA software moat that makes switching costs prohibitive for hyperscalers. While regulatory friction, as ChatGPT notes, is a valid concern, it acts as a barrier to entry that favors incumbents like MSFT and GOOG, who have the legal resources to navigate compliance. The real risk is not competition, but internal cannibalization of their own high-margin legacy software.
"Nvidia's moat is eroding from below (custom silicon) faster than regulatory friction protects it from above."
Gemini's CUDA moat argument is overstated. Yes, switching costs exist, but AMD's MI300 and custom silicon from hyperscalers (Google TPU, Meta Trainium) are eroding Nvidia's monopoly faster than the panel acknowledges. The real risk: if hyperscalers achieve 70%+ workload parity on in-house chips within 24 months, Nvidia's gross margins compress from 65% to mid-50s, invalidating the 22x multiple regardless of regulatory tailwinds. That's the cannibalization nobody's pricing.
"Hardware erosion alone won't compress Nvidia margins fast enough without macro or policy shocks hitting first."
Claude's 24-month parity timeline on custom silicon ignores that training workloads remain Nvidia-heavy even at 70% coverage, while inference margins could hold if CUDA lock-in persists beyond hardware swaps. This undercuts the margin-compression thesis and ties directly to ChatGPT's regulatory friction: export curbs would hit Nvidia first, accelerating any re-rating before alternatives scale.
The panelists agreed that AI is a significant growth opportunity but expressed concerns about over-optimistic valuations, regulatory risks, and potential cannibalization. They also debated the sustainability of Nvidia's hardware-software moat and the potential impact of custom silicon from hyperscalers.
AI monetization across ads, software, and data-center hardware was seen as a significant opportunity for growth.
Regulatory fragmentation and potential cannibalization of high-margin legacy software were the most frequently cited risks.