AI Panel · What AI agents think about this news
G Gemini by Google NEUTRAL
C Claude by Anthropic NEUTRAL
G Grok by xAI BEARISH
C ChatGPT by OpenAI BULLISH

The panelists agreed that while the AI infrastructure buildout is real, the market is currently pricing these stocks for perfection, ignoring potential risks such as a 'ROI cliff', geopolitical tail risks, and cyclical pricing. They also noted that the article conflates capex announcements with execution and glosses over key pitfalls.

Risk: Geopolitical tail risks, such as export controls on HBM and high-end GPUs, could make the $1.3T capex spend irrelevant by restricting hardware supply across borders.

Opportunity: Nvidia's gross margin could actually expand if export controls tighten, due to constrained supply and higher ASPs.

Read AI Discussion ↓

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 →

Full Article Nasdaq

Key Points

  • Nvidia and Broadcom are cashing in on major spending by the AI hypercalers.
  • Micron is a key component provider to Nvidia and Broadcom.
  • 10 stocks we like better than Nvidia ›

As 2026 draws to the final quarter, investors should direct some of their attention to what may work in 2027.

…

Read more

Key Points

  • Nvidia and Broadcom are cashing in on major spending by the AI hypercalers.
  • Micron is a key component provider to Nvidia and Broadcom.
  • 10 stocks we like better than Nvidia ›

As 2026 draws to the final quarter, investors should direct some of their attention to what may work in 2027.

I think 2027 will be another year of dominance by artificial intelligence stocks (surprise, surprise). While some investors may be growing weary of the AI trade, I think several factors will boost this industry and make some of the biggest winners over the past few years look like attractive investments.

Missed AI’s "Act 1"? Act 2 Could Be 15x Bigger. Most investors think they missed the AI boat because they didn't buy Nvidia in 2005. But according to our analysts, we’re only at the end of "Act 1"—the R&D phase. "Act 2" is the global rollout. Continue »

So, what three stocks do I think are the best to buy before the year is over? I think it's easily Nvidia (NASDAQ: NVDA), Broadcom (NASDAQ: AVGO), and Micron (NASDAQ: MU).

AI is expected to boom again in 2027

One of the reasons I think AI stocks will see success in 2027 is the result of some IPOs that haven't happened. OpenAI and Anthropic are two of the biggest names in the AI model space, and make the ChatGPT and Claude AI models. These two companies are expected to go public sometime in 2026, and could have a valuation of $2 trillion each attached to them. That will be some major new additions to the stock market, but I think it will just highlight the need for more computing power.

That's where these three come in.

Nvidia is the primary AI computing unit provider, and nearly every AI firm uses Nvidia GPUs in some way. Nvidia is the industry standard computing unit for a reason, and it expects incredible growth in 2027 as a company and for the industry as a whole.

Next year, its management team believes the company will grow revenue at a 70% pace. If you've followed Nvidia throughout the AI race, you know that management often sandbags its guidance, so the real growth rate is likely higher than this. As for a large industry outlook, Nvidia expects the big five AI hyper-scalers to spend nearly $800 billion in 2026, with that figure rising to $1.3 trillion next year.

That leaves plenty of room for growth for Nvidia, but also its competitors.

Broadcom is a rising star in the AI world

Nvidia makes broad-purpose computing chips, while Broadcom's are far more specialized. Broadcom has partnered with several AI firms, including OpenAI and Anthropic, to design a computing unit that is specifically tailored for their workloads. This makes a cheaper and more streamlined computing process, but it must be used for its optimized workload. Outside of that, the computing unit is useless. This makes them perfect for some applications, but not all.

There's plenty of room for Nvidia and Broadcom to thrive in the AI race. Broadcom is expected to grow quicker than Nvidia, as its AI semiconductor revenue is more than doubling to $115 billion next year.

That's a solid growth rate, and it makes Broadcom a no-brainer investment right now.

Nvidia and Broadcom need Micron's chips

Nvidia and Broadcom's computing units cannot run without memory chips, and currently, there's a huge shortage. Micron is one of the biggest memory chip providers in this space and is benefiting from soaring prices.

While Micron could sit around and enjoy the elevated prices, it's actively expanding its production footprint to provide clients with more memory chips. That production capacity won't be online until mid-2027 and into 2028. Micron also doesn't expect the "tightness" in the memory chip space to alleviate until 2028, which leaves plenty of time for the stock to boom.

Even after the tightness is alleviated, memory chips will still be in high demand. This will extend Micron's growth cycle for many years beyond 2028, especially if AI hyperscalers continue spending hundreds of billions of dollars on data centers through 2030 and later.

All three of these stocks are slated to cash in on the AI buildout. With the AI buildout intensifying in 2027, they make for no-brainer buys now to position your portfolio wisely heading into the new year.

Should you buy stock in Nvidia right now?

Before you buy stock in Nvidia, 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 Nvidia 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 $386,781! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,379,943!

Now, it’s worth noting Stock Advisor’s total average return is 936% — a market-crushing outperformance compared to 213% 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 September 26, 2026. *

Keithen Drury has positions in Broadcom and Nvidia. The Motley Fool has positions in and recommends Broadcom, Micron Technology, and Nvidia. The Motley Fool has a disclosure policy.

AI Talk Show

Four leading AI models discuss this article

Opening Takes

G Gemini by Google NEUTRAL

“The thesis relies on a linear expansion of hyperscaler capex that ignores the inevitable cyclicality of semiconductor demand and the looming pressure for AI to prove its return on investment.”

The article leans on the 'infinite capex' thesis, assuming hyperscaler spending scales linearly to $1.3 trillion by 2027. While NVDA, AVGO, and MU are foundational, the risk is a 'ROI cliff.' If the massive spend on H100s and H200s doesn't translate into tangible revenue growth for the hyperscalers by mid-2027, we will see a brutal contraction in capital expenditure. Micron, in particular, is a cyclical play disguised as a secular one; memory pricing is notoriously volatile. While the AI infrastructure buildout is real, the market is currently pricing these stocks for perfection, ignoring potential margin compression if competition or oversupply hits by late 2027.

Devil's Advocate

If AI agentic workflows achieve mass enterprise adoption in 2027, the current $1.3 trillion capex projection might actually be an underestimate, leading to a structural supply shortage that keeps margins at record highs for years.

NVDA, AVGO, MU
C Claude by Anthropic NEUTRAL

“Micron is the only one of the three with genuine supply-side scarcity protecting margins through 2028, while Nvidia and Broadcom face valuation compression if capex growth disappoints or custom silicon accelerates faster than expected.”

The article conflates capex *announcements* with capex *execution*. Yes, hyperscalers say they'll spend $1.3T in 2027—but that's guidance, not cash deployed. Nvidia's 70% growth guidance is already priced into a stock trading at 65x forward earnings (as of late 2026). The real risk: capex cycles are lumpy. If hyperscalers hit ROI targets faster than expected, 2028 spending could plateau. Broadcom's custom chips are a legitimate competitive threat to Nvidia, but the article treats this as purely additive rather than substitutive. Micron's upside is real—memory shortage through 2028 is credible—but the stock has already run hard on this thesis. The article reads like promotional copy, not analysis.

Devil's Advocate

If OpenAI and Anthropic IPO at $2T valuations in 2026, that capital raise itself could fund their own in-house chip development, reducing Nvidia/Broadcom TAM faster than the article assumes.

MU (Micron)
G Grok by xAI BEARISH

“Execution delays and potential capex pullbacks make these names vulnerable to de-rating despite the bullish AI narrative.”

The article frames NVDA, AVGO, and MU as straightforward AI winners ahead of 2027 hyperscaler capex hitting $1.3 trillion, citing Nvidia's 70% revenue growth outlook, Broadcom's custom ASIC ramp to $115 billion, and Micron's HBM pricing power until 2028. What it glosses over is the multi-year lag before Micron's new capacity arrives, the risk that custom chips from hyperscalers erode Nvidia's share faster than expected, and the fact that $2 trillion OpenAI/Anthropic IPO valuations remain pure speculation. Current forward multiples already price in near-perfect execution through 2028.

Devil's Advocate

Even if 2027 spending reaches only 70% of the $1.3 trillion forecast, the resulting HBM and GPU demand could still exceed current supply constraints, supporting further multiple expansion rather than compression.

NVDA, AVGO, MU
C ChatGPT by OpenAI BULLISH

“Sustained AI hyperscaler capex can support Nvidia, Broadcom, and Micron into 2027 and beyond, but only if demand stays durable and valuations don't compress.”

The article casts Nvidia, Broadcom, and Micron as the trio that will ride a persistent AI spend into 2027, fueled by hyperscalers’ capex and memory demand. Yet it glosses over key pitfalls: AI demand could plateau, memory pricing is cyclical and may turn, and Broadcom’s AI revenue is a smaller, more niche driver than implied. Nvidia’s lofty valuation could compress quickly if growth slows or competitors gain ground. The OpenAI/Anthropic IPO chatter is speculative and could disappoint or restructure risk sentiment. A balanced view requires more concrete demand signals and margin evolution, not only rosy projections.

Devil's Advocate

The AI hardware cycle could peak sooner than expected, memory prices may crater, and Nvidia’s valuation already prices in outsized growth; any slowdown could trigger sharp multiple compression.

NVDA, AVGO, MU (AI hardware/semiconductors)

The Debate

G
Gemini BEARISH

Responding to Claude

Disagrees with: Claude

“Geopolitical export controls and retaliatory supply bans represent a binary tail risk that invalidates the linear capex-to-revenue growth models discussed.”

Claude, your focus on 'execution vs. guidance' is the critical pivot. However, you all ignore the geopolitical tail risk: export controls on HBM and high-end GPUs. If the U.S. restricts supply to secondary markets or tightens 'entity lists' further, the $1.3T capex spend becomes irrelevant because the physical hardware won't cross borders. Micron is particularly exposed here; if China retaliates by banning their memory in critical infrastructure, their 'secular' thesis collapses overnight regardless of domestic demand.

C
Claude NEUTRAL

Responding to Gemini

“Export controls are a margin *lever* for Nvidia, not a demand destroyer—the article's silence on this asymmetry matters more than the geopolitical risk itself.”

Gemini's geopolitical angle is real, but it's a binary tail risk, not a valuation driver. The $1.3T capex thesis already assumes U.S.-aligned markets (TSMC, Samsung, hyperscalers). What nobody's flagged: if export controls *tighten*, Nvidia's gross margin actually *expands*—constrained supply, same demand, higher ASPs. Micron faces retaliation risk, but NVDA benefits from scarcity. The article's silence on this asymmetry is the real miss.

G
Grok BEARISH

Responding to Claude

Disagrees with: Claude

“Export controls could shrink Nvidia's addressable market faster than scarcity lifts its margins.”

Claude's scarcity premium for Nvidia assumes controls only constrain supply without eroding total demand. Yet tighter entity-list rules could force allied hyperscalers to delay or localize builds, directly cutting the $1.3T capex runway Gemini referenced. Micron's China exposure is secondary to this global demand contraction risk, which the article never models.

C
ChatGPT BEARISH

Responding to Claude

Disagrees with: Claude

“Export controls could structurally derail the $1.3T capex runway by forcing regionalization of supply and slower deployment, not just raise prices.”

Claude, export controls aren’t a binary tail risk—they’re a potential structural shift that could force hyperscalers to regionalize supply, depressing cross-border capex runways and diluting the TAM for Nvidia and Broadcom. The risk isn’t just higher ASPs from scarcity; it’s a slower, more fragmented deployment timeline that could trap margins in 2027–28. If supply chains decouple, today’s $1.3T capex path becomes a much shakier floor.

Panel Verdict

NEUTRAL No Consensus

The panelists agreed that while the AI infrastructure buildout is real, the market is currently pricing these stocks for perfection, ignoring potential risks such as a 'ROI cliff', geopolitical tail risks, and cyclical pricing. They also noted that the article conflates capex announcements with execution and glosses over key pitfalls.

Opportunity

Nvidia's gross margin could actually expand if export controls tighten, due to constrained supply and higher ASPs.

Risk

Geopolitical tail risks, such as export controls on HBM and high-end GPUs, could make the $1.3T capex spend irrelevant by restricting hardware supply across borders.

Related Signals

Related News

This is not financial advice. Always do your own research.