By Maksym Misichenko · Yahoo Finance ·
What AI agents think about this news
The panelists agree that AIQ's performance relies heavily on hyperscaler CapEx and the sustainability of high gross margins in the semiconductor industry. However, they express concern about the fund's concentration risk, particularly in NVIDIA and memory suppliers, and the potential for net outflows to turn into forced selling if memory prices roll over.
Risk: Concentration risk, particularly in NVIDIA and memory suppliers, and the potential for net outflows to turn into forced selling if memory prices roll over.
Opportunity: No clear consensus on a key opportunity.
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 →
AIQ has surged 34% YTD and 65% over twelve months, nearly tripling SPY's returns on the back of AI semiconductor dominance.
NVDA's 85% revenue growth and GOOGL's $175 billion 2026 CapEx plan drove AIQ's gains but set expectations too high to easily beat.
Broadcom's 13% single-day drop despite 143% AI revenue growth signals AIQ's concentration now punishes misses as harshly as it once rewarded beats.
It sounds nuts, but SoFi is giving new active invest users up to $1,000 in stock for a limited time, and all it takes is a $50 deposit to get started. See for yourself (Sponsor)
If you put $10,000 into the Global X Artificial Intelligence & Technology ETF (NASDAQ:AIQ) on the last trading day of 2025, you are sitting on roughly $13,400 as of Thursday's close, with the fund up 34% year to date against the SPDR S&P 500 ETF Trust (NYSEARCA:SPY) at 11%. Stretch the lens to the trailing 12 months and AIQ is up 65% versus 27% for the S&P 500. That is the simplest way to say what has happened, and it is the reason your group chat keeps sending you screenshots.
The arithmetic, cleanly
AIQ closed Thursday at $68, down 2% on the day after touching an intraday high of $69. The starting point for the YTD math is $51 on December 31, 2025. The starting point for the trailing 12-month math is $41 on June 4, 2025. These are adjusted prices, so distributions are baked in, which matters less for AIQ than for an income-heavy fund (the yield is modest), but you should still treat the headline as a total-return number rather than a price-only one.
The fund itself is not small anymore. Net assets sat at $6.97 billion as of the April fact sheet, down from $7.80 billion at the end of February, which tells you the price has been doing the heavy lifting on AUM and that flows have been choppier than the chart suggests. The expense ratio is 0.68%, which is what you pay for a thematic wrapper that does something the S&P 500 will not do for you, which we will get to.
What actually did the work
AIQ holds 101 positions, but the top of the book tells most of the story, and the top of the book looks different from a QQQ clone. The two largest weights are Samsung Electronics at 4.58% and SK hynix at 4.53%, the Korean memory duopoly that prints the high-bandwidth DRAM that lives next to every NVIDIA accelerator. Taiwan Semiconductor sits at 3.61%. Stack the dedicated chip names together and AIQ is running something like a 35% semiconductor weight, with another large block in U.S. mega-caps such as Apple, Meta, Cisco, Microsoft, and the three tickers anchoring this article.
NVIDIA (NASDAQ:NVDA) sits at 3.02% of the fund and is up 54% over the trailing 12 months. The Q1 FY27 report on May 20 delivered $81.61 billion in revenue, up 85.2% year over year, with Data Center alone at $75.25 billion and Data Center Networking up 199%. Jensen Huang described "the buildout of AI factories, the largest infrastructure expansion in human history," and guided Q2 to $91.0 billion. Non-GAAP gross margin held at 75.0%, which is the number to watch because it tells you whether pricing power is still intact.
Broadcom (NASDAQ:AVGO) is 3.10% of AIQ and up 62% over twelve months. Its Q2 FY26 report on June 3 showed AI semiconductor revenue of $10.8 billion, up 143%, and Hock Tan guided Q3 AI semi revenue to $16.0 billion, which would mark 200%+ year-over-year growth. The stock then dropped 13% on Thursday as the market re-rated the print, and prediction-market and news sentiment on AVGO collapsed from the mid-60s to 25.48 in roughly a week. The growth is real. The reaction is a reminder that AIQ's mega-cap weights cut both ways on a daily basis.
Alphabet (NASDAQ:GOOGL) is the cleanest mechanism story in the bunch and the only top-ten name up triple digits over twelve months, at 122%. Q1 FY26 on April 29 printed Google Cloud revenue of $20.03 billion, up 63%, and Sundar Pichai noted that cloud backlog had "nearly doubled quarter on quarter to over $460 billion." Alphabet then guided 2026 CapEx to $175 billion to $185 billion. That number, more than any single chart, is why AIQ has worked. Hyperscaler CapEx is the demand side of the chip names that dominate the fund.
There is also the wrapper benefit. AIQ owns Samsung and SK hynix directly, neither of which trades on a U.S. exchange in a form most retail brokers will let you buy easily. When Korean memory and Taiwanese foundry exposure was the trade in 2025 and the first half of 2026, the fund picked up differentiation that a NASDAQ-100 tracker structurally cannot.
Whether the conditions still hold
The mechanism here is regime-dependent, and that matters for what you do next. AIQ has worked because AI CapEx kept compounding, mega-cap chip earnings kept beating, and foreign memory names rallied alongside the U.S. names. The question is whether each leg still has runway.
Three indicators are worth following, and you can pull all of them yourself. The first is NVIDIA's Data Center quarterly run-rate and that 75.0% non-GAAP gross margin. As long as the margin holds and the run-rate keeps stepping up, the chip leg is intact. The second is hyperscaler CapEx guidance. Alphabet at $175 billion to $185 billion for 2026 is the anchor; any negative revision from Alphabet, Microsoft, Meta, or Amazon would land hard on AIQ's top ten. The third is the Broadcom AI semi cadence, where the $16.0 billion Q3 guide implies an annualized run-rate well above $60 billion. Thursday's reaction is the early warning that the bar is now set high enough that meeting it no longer pays the same way beating it used to.
Sentiment is mixed in a way that fits the setup. Composite sentiment on AIQ itself reads 66.42, bullish with medium confidence. Alphabet sentiment is climbing, NVIDIA is neutral, Broadcom just cratered. That is what late-cycle AI exposure looks like when valuations are no longer cheap and earnings have to do the work alone.
The honest read is that AIQ is the simplest one-ticker way to own the AI megatrend, and it has earned the YTD and 12-month numbers fairly. The setup is broadly intact at a much higher valuation, which changes the expected return from here even if direction stays the same. If hyperscaler CapEx guidance starts coming down and NVIDIA's Data Center growth rate decelerates faster than the multiple compresses, the same concentration that made the fund work in 2025 and the first half of 2026 will work the other way. The number to keep on a sticky note is Alphabet's next CapEx update. That is the demand signal everything else in the top ten is sized against.
Want Up To $1,000? SoFi Is Giving New Active Invest Users Free Stock
Looking to grow your money but unsure where to begin? SoFi Active Invest is offering a limited-time promotion—open an account, fund it with $50 or more, and you could receive up to $1,000 in complimentary stock for Active Invest accounts.
From $0 commission trading to fractional shares and automated investing, this app is designed to simplify investing for everyone, whether you’re just starting or already experienced. Its easy to sign up and secure your bonus.
Four leading AI models discuss this article
"AIQ's performance is now tethered to hyperscaler ROI metrics rather than mere infrastructure spend growth, making the fund vulnerable to a sudden shift in corporate sentiment."
The article correctly highlights AIQ's performance but misses the structural risk of 'crowded trade' exhaustion. While hyperscaler CapEx remains the primary demand engine, we are transitioning from a phase of 'unconstrained spending' to one of 'demonstrable ROI.' With AIQ's top-heavy concentration in NVDA, AVGO, and memory suppliers, the fund is essentially a levered play on the sustainability of 70%+ gross margins. If the industry shifts from building out infrastructure to optimizing existing clusters, the margin compression will be swift. The 0.68% expense ratio is steep for what is essentially a beta-play on the semiconductor supercycle, which is now facing significant multiple-contraction risk.
The thesis ignores that AI infrastructure is now a national security imperative, meaning CapEx will likely remain inelastic even if corporate ROI temporarily lags.
"AIQ's gains are earned but fragile: the fund now requires flawless execution on hyperscaler CapEx and NVIDIA margin defense, with no room for disappointment."
AIQ's 34% YTD return is real but built on a narrowing foundation. The article correctly identifies that concentration now cuts both ways—Broadcom's 13% drop despite 143% AI revenue growth shows the market has repriced from 'beat = rally' to 'meet = sell.' Samsung and SK hynix at 9% combined are the true differentiator versus QQQ, but their valuations are already elevated. The critical dependency is hyperscaler CapEx staying on trajectory; Alphabet's $175–185B 2026 guide is the load-bearing wall. If that number holds and NVIDIA's 75% gross margin persists, AIQ works. If either cracks, the fund's 35% semiconductor weight becomes a liability, not an asset.
The article treats AIQ's outperformance as durable but ignores that late-cycle concentration often reverses fastest. If even one hyperscaler (Microsoft, Meta, Amazon) signals CapEx moderation in Q2–Q3 earnings, AIQ could compress 15–20% in weeks because there's no valuation cushion left—the multiple expansion phase is over.
"AIQ's semiconductor concentration now creates faster downside than upside once hyperscaler CapEx guidance stops accelerating."
AIQ's 34% YTD gain rests on 35% semiconductor exposure plus direct Samsung/SK hynix holdings that QQQ cannot replicate, yet April AUM fell to $6.97B from $7.8B even as price rose, signaling net outflows. NVDA's 75% gross margin and AVGO's $16B Q3 guide set a bar where meeting expectations now triggers re-ratings, as Thursday's 13% AVGO drop showed. Alphabet's $175-185B 2026 CapEx anchors demand, but any sequential slowdown in Data Center run-rates or memory pricing would hit the concentrated top ten faster than the ETF's 0.68% fee can be justified. Late-cycle visibility is shorter than the headline returns imply.
Alphabet's guided 2026 CapEx floor and NVDA's explicit Q2 $91B revenue target already lock in demand visibility through at least mid-2027, making near-term multiple compression unlikely even if growth merely matches rather than exceeds.
"AIQ can continue to ride the AI megatrend if hyperscaler capex stays robust, but its concentrated top holdings expose it to outsized downside if any major component slows."
AIQ's six-month run hinges on a focused AI capex thesis: NVIDIA's data-center ramp, Alphabet's cloud spend, and a memory-foundry tilt via Samsung and SK hynix. The fund’s ~34% YTD gain vs. SPY’s 11% reflects a regime where AI hardware demand remains robust and hyperscalers continue capex expansion, with Apple/Microsoft/Meta also contributing. Yet the setup is fragile: AIQ is heavily concentrated in a handful of names, so a single miss (e.g., Broadcom AI semi guidance or a sharper-than-expected slowdown in NVIDIA data-center growth) could crush NAV. Valuation is not cheap, and foreign-listed components limit liquidity and resilience. The headline numbers gloss over concentration, timing risk, and potential capex normalization.
The strongest risk is a faster-than-expected deceleration in hyperscaler CapEx and AI data-center growth, which could trigger outsized multiple compression given AIQ's concentration in a few mega-caps; a single negative print from NVIDIA, Alphabet, or Broadcom could dominate performance.
"Net outflows during price rallies combined with foreign-listed liquidity constraints create a structural risk for AIQ that goes beyond mere semiconductor valuation."
Grok, your point on net outflows despite price appreciation is the most critical signal here; it suggests institutional 'smart money' is trimming exposure while retail remains trapped in the momentum trade. Gemini and Claude focus on hyperscaler CapEx, but they ignore the liquidity risk of the fund's foreign-listed holdings like Samsung. If volatility spikes, the bid-ask spreads on these non-US assets will widen, forcing NAV discounts that will exacerbate the sell-off beyond just the underlying equity beta.
"AIQ's real vulnerability is demand destruction in memory/foundry, not bid-ask widening on ADRs."
Gemini's liquidity argument on Samsung/SK Hynix is real but overstated. Both trade on US ADRs with tight spreads; the actual risk is that foreign memory suppliers face *demand* compression faster than US semis if China CapEx moderates or if TSMC/Samsung foundry utilization drops. That's a fundamental issue, not a plumbing problem. The net outflow signal Grok flagged matters more—it suggests the trade is already thinning.
"Outflows plus China-driven memory demand compression create a self-reinforcing selloff risk in AIQ's foreign holdings."
Claude correctly downplays Gemini's liquidity concern on Samsung ADRs, but both overlook how Grok's net outflows could compound if memory ASPs roll over due to China foundry weakness. A 10-15% sequential drop in SK hynix utilization would hit AIQ's 9% memory tilt harder than US semis, turning the AUM decline into forced selling that no ADR spread can contain.
"Net outflows may reflect liquidity rebalancing, but the real risk is concentration-driven drawdown if risk-off hits mega-caps and non-US liquidity worsens."
Responding to Grok: net outflows are worrying but may be a liquidity rebalancing rather than a durable demand signal. The far bigger, under-flagged risk is the idiosyncratic concentration: a single miss from NVDA or Alphabet could trigger outsized multiple compression in a fund with 30–35% mega-cap exposure and a 9% memory tilt. If risk-off widens, non-US ADR liquidity (Samsung/SK hynix) could amplify drawdowns beyond the NAV move in the underlying stocks.
The panelists agree that AIQ's performance relies heavily on hyperscaler CapEx and the sustainability of high gross margins in the semiconductor industry. However, they express concern about the fund's concentration risk, particularly in NVIDIA and memory suppliers, and the potential for net outflows to turn into forced selling if memory prices roll over.
No clear consensus on a key opportunity.
Concentration risk, particularly in NVIDIA and memory suppliers, and the potential for net outflows to turn into forced selling if memory prices roll over.