BlackRock’s iShares Future AI & Tech ETF has lagged far behind the S&P 500 this year. (Michael M. Santiago/Getty Images)

Weird Money

The best AI fund of 2024? The S&P 500.

High-fee AI ETFs are great for asset managers, but not so good for investors.

Jack Raines

9/3/24 2:24PM

If I asked you to name the defining technological trend of the past two years, you would probably say, “artificial intelligence,” and if I asked you how artificial intelligence stocks had performed over the last two years, you would probably say, “Pretty well!” Even after its recent sell-off, Nvidia is up ~900% since Fall 2022, SMCI is up ~700%, Meta has tripled, and Microsoft has gained roughly 80%. And yet, according to The Wall Street Journal’s James Mackintosh, every AI-themed ETF has underperformed the S&P 500:

Pity the investors in the three artificial-intelligence-themed exchange-traded funds that managed to lose money this year. Every other AI-flavored ETF I can find has trailed both the S&P 500 and MSCI World. That is before the AI theme itself was seriously questioned last week, when investor doubts about the price of leading AI stocks Nvidia and Super Micro Computer became obvious.
The AI fund disaster should be a cautionary tale for buyers of thematic ETFs, which now cover virtually anything you can think of, including Californian carbon permits (down 15% this year), Chinese cloud computing (down 21%) and pet care (up 10%). Put simply: You probably won’t get what you want, you’ll likely buy at the wrong time and it will be hard to hold for the long term.
Ironically enough, Nvidia’s success has made it harder for some of the AI funds to beat the wider market. Part of the point of using a fund is to diversify, so many funds weight their holdings equally or cap the maximum size of any one stock. With Nvidia making up more than 6% of the S&P 500, that led some AI funds to have less exposure to the biggest AI stock than you would get in a broad index fund.

How have so many artificial intelligence funds underperformed the S&P 500? Well, for starters, the S&P is top-heavy with some of the biggest current winners of the AI boom: its six largest components, which make up 28% of the index, are Apple, Microsoft, Nvidia, Amazon, Meta, and Alphabet. Meanwhile, the six largest positions in BlackRock’s iShares Future AI & Tech ETF are Broadcom, Nvidia, AMD, Palantir, Fortinet, and Accenture. While I do appreciate BlackRock including Accenture, a management consulting firm with $3.6 billion in annualized generative AI bookings, in its AI ETF, it’s surprising that the asset manager weighted it heavier than Amazon, Microsoft, Alphabet, and Taiwan Semiconductor.

The issue at hand is that betting on market trends and betting on individual companies are two very, very different endeavors. An association with “AI” doesn’t guarantee that a company’s stock will benefit from AI, at least not in the long-run. AI has to, at some point, translate to cash flow for the business. Compounding this issue is the fact that “trend” winners might be concentrated, but ETFs tend to be diversified. Nvidia’s market cap may have increased by 900% since Fall 2022, but if a fund has a max position size mandate, it will be forced to diversify into worse-performing companies (such as, you know, Accenture and Intel).

Imagine, for example, that you invested in an “electric vehicle” ETF in 2022 that was equal-weighted to Tesla, Fisker, Nio, Nikola, Canoo, Lucid Motors, and Rivian. While Tesla has been roughly flat over that time, the other companies are down significantly. Increasing electric vehicle adoption did not necessarily mean that all electric vehicle stocks would do well. The businesses themselves matter.

So why, given the underperformance, do so many asset managers issue thematic ETFs? Because they can charge hefty fees and expenses. BlackRock’s iShares Future AI & Tech ETF charges 0.47%, while the expense ratio on its S&P 500 ETF is just 0.03%. Thematic ETFs are lucrative for asset managers, regardless of how their investors fare. If you want to play the AI trend (or any market trend, for that matter), it’s probably best to either do your own due diligence on winners and losers or simply stick with index funds.

Max Knoblauch3h

Airlines rise, continuing their volatile 2026, as US-Iran talks may foreshadow some oil supply relief

Airline stocks are surging on Friday, as the market appears to be pricing in some medium-term oil pricing relief following talks between the US and Iran. Iranian officials referred to the meeting as “a good beginning.”

Shares of budget carriers, which have tighter margins and are more sensitive to fluctuations in fuel costs, are leading the surge. Frontier Airlines and Allegiant up more than 13%, while major airlines like United Airlines, American Airlines, and Delta Air Lines are also up at least 6%. JetBlue and Alaska Air are similarly up about 6%.

The market more broadly is rebounding on Friday, with the S&P 500 up 1.6% and bitcoin recovering some of this week’s losses.

Airlines have been volatile to start 2026 amid geopolitical tensions, varying annual forecasts, and the impact of winter storms.

Luke Kawa4h

The AI supply chain is soaring thanks to Amazon’s capex budget

If tech companies are going to spend way more than expected on capex, well, that means other companies are poised to benefit from that massive spending spree.

Amazon’s plan for $200 billion in business investment this year was the exclamation point to end a reporting period that saw every Magnificent 7 hyperscaler that provides guidance offer a 2026 capex budget well above what Wall Street had anticipated.

Here’s a look at the different parts of the supply chain that are soaring on the persistent demand for, and seeming scarcity of, AI compute:

Data center/neocloud companies: Applied Digital, CoreWeave, Cipher Mining, and Nebius, with some of these seemingly getting an extra lift from their legacy role as crypto miners.
Chips: the usual suspects — Nvidia, Broadcom, and Advanced Micro Devices.
Racks: Super Micro Computer, which is continuing its strong run following earnings, and Dell.
Power: Oklo, Nuscale, and Plug Power.

Big Tech capital expenditure soared in 2025. It’s going up another 50% in 2026.

Here’s a look at the different parts of the supply chain that are soaring on the persistent demand for, and seeming scarcity of, AI compute:

Data center/neocloud companies: Applied Digital, CoreWeave, Cipher Mining, and Nebius, with some of these seemingly getting an extra lift from their legacy role as crypto miners.
Chips: the usual suspects — Nvidia, Broadcom, and Advanced Micro Devices.
Racks: Super Micro Computer, which is continuing its strong run following earnings, and Dell.
Power: Oklo, Nuscale, and Plug Power.

Matt Phillips5h

For memory chips, the “parabolic price hike” is continuing to ramp higher

The remarkable run-up in prices for memory chips continued into early February, analysts at Bernstein Research say, driven largely by data center demand from hyperscalers and cloud service providers (CSP).

Prices for NAND flash memory wafers — a type of memory used in devices, as it retains data even when powered down — soared 35% between the end of 2025 and February 2.

Spot prices for DRAM — ubiquitous short-term data storage chips — jumped about 28% in that period. But that massively understates the remarkable shift in pricing for what were long seen as commodity tech hardware inputs. DRAM prices are more than 2,000% over the last year, while NAND prices are up more than 600% in that period.

The ongoing momentum provides still more support for memory chip plays like Micron and Sandisk, which have been big market winners in recent months.

In a note published earlier this week, Bernstein Research analysts wrote:

“The parabolic price hike continued in Jan. Indicated price increase for 1QCY26 is much stronger than we expected and we hence see upside to our near term memory pricing projection. Unrelenting CSP demand remained the main driver. PC and Mobile demand hasn’t been destroyed yet because of lean inventory & pull-forward purchase. Going forward price hike is expected to continue but likely at a slower rate, as PC and Mobile demand should contract meaningfully this year. Price however may stay elevated throughout this year, supported by CSP demand.”

Chip Unveils Rap Star Wax Figures At Madame Tussauds

Why there’s a “huge vibe divergence” between tech and finance on AI

Tech evangelists are hailing a Claude-fueled seismic shift in computer-based work. Investors are, by and large, selling AI stocks.

Luke Kawa5h

Matt Phillips6h

Bloom Energy earnings get warm reception from analysts

Fuel cell-based power provider Bloom Energy posted better-than-expected Q4 earnings and sales results after the bell on Thursday, sending the stock higher aftermarket and into early Friday trading. Here’s some of the positive chatter from analysts reacting to the bullish results:

Barclays: “What to know: 1) 2026 guide well above the Street for all metrics; 2) Product backlog comes in at $6.0bn with services backlog of $14.0 bn, reflecting 100% attach rate on new bookings.”

Morgan Stanley: “An inflection in growth is now beginning to show up in the financials. Significant 4Q25 earnings beat, product backlog up 2.5x, and 2026 revenue guidance meeting our Street-high forecast: >50% YoY as demand begins to ramp. We stay OW, raise PT to $184 on recent project wins.”

JP Morgan: “We are adjusting our estimates and introducing FY28 estimates with this note. Our YE26 price target goes to $166, from $154. While the stock has significantly outperformed YTD, we maintain our Overweight rating and believe that additional contract announcements should provide further positive catalysts and potentially increased visibility into our unit shipment vs margin sensitivity analysis (see below).”

Evercore ISI: “The most noticeable and arguably most anticipated metric Bloom provided was its current product backlog which currently stands at $6B representing a ~2.5x increase YoY, with total current backlog (product and services) ballooning to $20B. These impressive backlog metrics should provide confidence in the company’s ability to deliver on its newly established $3.1-$3.1B 2026 revenue target (vs. cons. of ~$2.1B) and double its non-GAAP operating income ($450M midpoint vs. $221M 2025A).”