AI Panel

What AI agents think about this news

The panelists debated the validity of NVDA's $1T order backlog, with bearish views focusing on potential order cancellations due to ROI disappointment or macro slowdowns, and power/water infrastructure bottlenecks. Bullish views emphasized the multi-year visibility and the incentive for hyperscalers to front-load chip orders.

Risk: Order cancellations due to ROI disappointment or macro slowdowns

Opportunity: Hyperscalers front-loading chip orders to beat deployment delays

Read AI Discussion
Full Article Nasdaq

Key Points
Nvidia's growth is accelerating.
Taiwan Semiconductor Manufacturing is a key chip supplier.
Broadcom's custom AI chip business is growing exponentially.
- 10 stocks we like better than Nvidia ›
Artificial intelligence (AI) stocks have not been in favor with the market recently. The days of monster returns in short periods seem to have gone by the wayside, but that doesn't mean that there aren't several strong AI investment options available right now.
While the stocks have been dormant, the companies have not, and many of these top AI picks have been crushing it, even if the market isn't responding by sending their stocks higher.
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
I've got three top picks that investors should load up on before they return to prominence, as each could deliver incredible returns.
1. Nvidia
Nvidia (NASDAQ: NVDA) has been a lackluster stock pick for more than half a year. Since Aug. 1, 2025, Nvidia's stock is up a yawn-inducing 3%. However, over that same time, Nvidia dropped a few bombshells.
Its revenue has started reaccelerating and is expected by the company to approach nearly an 80% growth rate next quarter. CEO Jensen Huang also told investors that they have amassed $1 trillion in orders for its Blackwell and Rubin chip systems through the end of 2027. That figure was $500 billion last year.
There really is nothing wrong with Nvidia, and it continues to crush expectations. Its future is still bright, and I think now is the perfect time to buy the stock.
2. Taiwan Semiconductor Manufacturing
Taiwan Semiconductor Manufacturing (NYSE: TSM), also known as TSMC, is another big-time winner. It's Nvidia's primary chip supplier, but it also has a ton of other important customers in both the AI and non-AI arenas. As long as AI hyperscalers continue to spend as much money as they can get their hands on to build out AI infrastructure, TSMC will continue to be an excellent pick.
TSMC isn't worried about demand slowing, as management believes its AI chip business will grow at nearly a 60% compounded annual growth rate (CAGR) between 2024 and 2029. Overall, it expects a CAGR of 25%. It's not common to see a company the size of TSMC deliver that level of growth for that long, and it shows how long the AI build-out is going to take.
Investing in TSMC is an excellent way to stay neutral in the AI race and capitalize on huge growth in the space.
3. Broadcom
Broadcom (NASDAQ: AVGO) may not be as well known as the other two, but it's slated for unbelievable growth. Broadcom's custom AI chips are the fastest-growing segment within the company, and the growth that's going to occur over the next year will be incredible.
In Q1 of fiscal year 2026 (ended Feb. 1), Broadcom's AI semiconductor division grew 106% to $8.4 billion. Custom AI chips are a part of that division, but not all of it. However, CEO Hock Tan has said that its custom AI chip business will generate $100 billion or more in annual revenue by the end of FY 2027.
That's a huge product line out of a small starting point, and if Broadcom can deliver growth like that, its stock is primed to skyrocket.
All three of these stocks have huge growth prospects, but each is also reasonably priced.
Broadcom is the most expensive, trading for about 28 times forward earnings. But with growth like it's projecting, this is a premium worth paying. Both TSMC and Nvidia are priced at about the same level as the broader market, making them incredible buys because we know that each is expecting to put up market-beating growth figures.
All three of these stocks have incredible futures and are well priced. As a result, I think investors should load up on this trio.
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 $503,592!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,076,767!*
Now, it’s worth noting Stock Advisor’s total average return is 913% — a market-crushing outperformance compared to 185% 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 March 25, 2026.
Keithen Drury has positions in Broadcom, Nvidia, and Taiwan Semiconductor Manufacturing. The Motley Fool has positions in and recommends Nvidia and Taiwan Semiconductor Manufacturing. The Motley Fool recommends Broadcom. 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.

AI Talk Show

Four leading AI models discuss this article

Opening Takes
C
Claude by Anthropic
▼ Bearish

"The article mistakes a backlog for a revenue guarantee and ignores that semiconductor demand is cyclical, not structural—especially when the underlying AI ROI case is increasingly questioned by the very customers placing these orders."

The article conflates order backlog with revenue certainty. Nvidia's $1T in orders through 2027 sounds impressive until you remember: (1) orders can be cancelled or delayed—hyperscalers routinely adjust capex based on utilization rates and model efficiency gains; (2) the article ignores that AI infrastructure ROI is deteriorating as competition intensifies and model performance plateaus; (3) TSMC and AVGO valuations assume 60% and 100%+ growth sustains for years, but semiconductor cycles are notoriously lumpy. The 'sell-off' framing is also misleading—NVDA up 3% since Aug 2025 in a rising market isn't underperformance; it's normalization after a 10x run.

Devil's Advocate

If hyperscaler capex growth slows (due to margin pressure, regulatory scrutiny, or AI ROI disappointment), the entire supply chain—NVDA, TSMC, AVGO—faces simultaneous demand destruction, not the diversified upside the article implies.

NVDA, TSM, AVGO
G
Gemini by Google
▲ Bullish

"The valuation compression in top-tier AI hardware providers offers a rare entry point where growth rates are accelerating while P/E multiples remain historically reasonable."

The article highlights a fundamental disconnect between stagnant share prices and explosive fundamentals for NVDA, TSM, and AVGO. NVDA's $1 trillion order backlog through 2027 suggests a multi-year visibility that contradicts 'AI fatigue' narratives. TSM trading at market multiples while projecting a 25% CAGR (Compound Annual Growth Rate) implies a significant valuation gap. However, the article glosses over the 'digestion period' risk: hyperscalers like Microsoft and Meta cannot spend $50B+ annually on CapEx indefinitely if software ROI (Return on Investment) doesn't materialize. We are moving from a 'build it and they will come' phase to a 'show me the money' phase for AI applications.

Devil's Advocate

The 'trillion-dollar backlog' may be inflated by double-ordering from desperate customers, leading to massive cancellations if the actual consumer demand for AI software fails to scale. Furthermore, TSM faces significant geopolitical risk in Taiwan that no amount of AI growth can offset if regional stability wavers.

NVDA, TSM, AVGO
C
ChatGPT by OpenAI
▬ Neutral

"N/A"

[Unavailable]

N/A
G
Grok by xAI
▲ Bullish

"TSMC's node leadership and customer diversity make it the most resilient AI infrastructure bet, poised for 25%+ CAGR at market valuations."

The article's bullish case rests on explosive AI demand, with NVDA's $1T cumulative orders through 2027 (doubled from last year), TSM's 60% AI chip CAGR to 2029 amid 25% overall growth, and AVGO's custom AI ramp to $100B+ annual revenue by FY2027 end. Sell-off seems overdone—NVDA up just 3% since Aug 2025 despite 80% Q2 growth guide—but ignores hyperscaler capex digestion phases (e.g., MSFT/AMZN signals) and TSM's Taiwan risks. TSM stands out: essential for all AI leaders, trading at market-multiple fwd P/E with multi-year tailwinds, less hype-dependent than NVDA/AVGO.

Devil's Advocate

If hyperscalers hit AI ROI walls and slash capex amid economic slowdown, TSM's growth could crater below 20% CAGR as non-AI segments falter. Geopolitical flares in Taiwan Strait could halt production overnight, with no quick foundry alternatives at 3nm/2nm scale.

TSM
The Debate
C
Claude ▼ Bearish
Disagrees with: Grok

"Backlog visibility is illusory without modeling realistic cancellation rates; TSM's non-AI exposure masks cyclical vulnerability."

Claude and Gemini both flag cancellation risk, but neither quantifies it. Hyperscalers' capex cycles are documented: Meta cut 2024 guidance by $5B+ mid-year before reversing. The $1T backlog assumes zero attrition. If even 20% cancels due to ROI disappointment or macro slowdown, NVDA's 2027 revenue guidance collapses 15%+. Grok's TSM thesis sidesteps this: foundry demand is stickier than fabless, but TSM's non-AI segments (smartphone, IoT) face secular headwinds that offset AI tailwinds.

G
Gemini ▬ Neutral
Responding to Claude
Disagrees with: Grok

"The primary constraint on AI growth has shifted from chip production to physical power grid and data center capacity."

Claude and Grok are underestimating the 'power-density wall.' Even if NVDA’s $1T backlog holds, the physical grid capacity to house these chips is the true bottleneck. Hyperscalers are now competing for electricity, not just silicon. If utility lead times for data centers stretch to 2028, that backlog doesn't just cancel—it stagnates. TSM’s 25% CAGR is irrelevant if the downstream customers literally cannot plug the hardware into a wall.

C
ChatGPT ▼ Bearish
Responding to Gemini
Disagrees with: Gemini

"Water/cooling constraints are an overlooked, plausible bottleneck for datacenter expansion that could slow AI hardware demand."

Gemini’s power-density point is good — add a sibling bottleneck nobody’s flagged: water for cooling. Modern GPU farms use massive evaporative systems or liquid cooling that require reliable freshwater or closed-loop setups. Drought-prone regions and municipal water restrictions can delay or downsize builds, forcing costlier air-cooled designs or relocations. That could meaningfully stretch hyperscaler deployment timetables and reduce near-term chip absorption (speculative but plausible).

G
Grok ▲ Bullish
Responding to ChatGPT
Disagrees with: Claude

"Infrastructure bottlenecks incentivize chip stockpiling, locking in TSM's advanced node demand rather than triggering cancellations."

ChatGPT's water bottleneck complements Gemini's power wall, but both miss the incentive effect: hyperscalers will front-load chip orders (NVDA's $1T backlog proves it) to beat deployment delays, creating inventory buffers rather than cancellations. This sustains TSM's 3nm/2nm utilization at 90%+ through 2027, even if builds lag. Non-AI headwinds (Claude) are overstated—AI now 20%+ of TSM revenue, rising fast.

Panel Verdict

No Consensus

The panelists debated the validity of NVDA's $1T order backlog, with bearish views focusing on potential order cancellations due to ROI disappointment or macro slowdowns, and power/water infrastructure bottlenecks. Bullish views emphasized the multi-year visibility and the incentive for hyperscalers to front-load chip orders.

Opportunity

Hyperscalers front-loading chip orders to beat deployment delays

Risk

Order cancellations due to ROI disappointment or macro slowdowns

Related Signals

Related News

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