BlackRock says the artificial intelligence rally gripping U.S. markets is powerful, expensive and increasingly concentrated, but it is not a repeat of the late-1990s internet bubble in the way it is being financed or in the strength of corporate profits behind it.
In its latest weekly commentary, the asset manager argued that the main question is not whether AI-linked stocks have risen too far too fast, but whether companies can keep delivering the profit growth now expected by the market. BlackRock said the label “bubble” carries a major assumption: that AI will fail to produce lasting productivity gains large enough to justify today’s spending and valuations.
The comparison with the dot-com era is unavoidable. Technology stocks have surged since 2019, market concentration has climbed above levels seen at the peak of the internet boom, and broad valuation measures are again near historic highs. But BlackRock said the structure of the current rally looks different. Today’s largest technology companies are mostly profitable, generate substantial free cash flow and are funding AI-related capital spending largely from internal resources rather than debt.
That distinction has become central to the debate across equity, bond and digital-asset markets. If AI spending continues to translate into earnings growth, high valuations may be easier to defend. If profit expectations begin to crack, the same concentration that has lifted U.S. equity indexes could magnify losses across risk assets, including some cryptocurrency sectors tied to computing, storage and data infrastructure.
The central question is earnings
BlackRock framed the AI debate around corporate earnings rather than price action alone. According to data cited in the commentary, technology stocks are up 569% since 2019. That is a dramatic gain, far ahead of the broader U.S. equity market, which advanced 237% over the same period.
Still, the move remains smaller than the internet-era surge. From 1993 to 1999, technology shares climbed 1,097%, including back-to-back annual returns above 78% in the final two years of that decade. The late-1990s rally was also marked by a flood of new companies with little or no profit, aggressive borrowing and extremely optimistic forecasts about internet adoption.
The current cycle has been different in one critical way: profits have already risen sharply for many of the largest technology companies. S&P 500 companies are expected to post a 23% year-on-year profit increase in the second quarter of 2026, which would mark a seventh consecutive quarter of double-digit earnings growth. Large technology groups are expected to grow earnings by more than 30%, according to data cited by BlackRock, while trading at an aggregate valuation near 26 times earnings.
That combination explains why the rally has been difficult to dismiss. Prices are high, but earnings are also rising quickly. The market’s challenge is that both sides of that equation must continue working. If earnings growth slows while valuations remain stretched, prices may become more vulnerable.
Valuations are stretched, but not all measures say the same thing
The warning signs are clear. Morningstar data cited by BlackRock show that technology stocks now account for 37.5% of total U.S. equity market capitalization. That is higher than the concentration seen at the peak of the dot-com era.
The S&P 500’s Shiller price-to-earnings ratio, which smooths earnings over a 10-year period, has returned to around 40 times. That matches the level reached in 2000, just before the internet bubble burst and technology shares entered a long downturn.
At the same time, the index’s 12-month forward price-to-earnings ratio stands at 21 times. That is elevated by historical standards, but not as extreme as the Shiller measure. The gap reflects stronger near-term earnings expectations. In plain terms, traders are paying high prices, but they are also expecting much stronger profits in the next year.
This is why the market debate has become more nuanced than a simple bubble call. A high Shiller ratio suggests long-term valuations are stretched. A lower forward multiple suggests current earnings momentum is providing support. Both can be true at the same time.
BlackRock’s view is that the next phase depends on whether earnings can catch up with prices. If AI spending produces large revenue gains, lower costs or broader productivity improvements, current valuations may prove durable. If AI remains costly while returns take longer than expected, the market could reprice quickly.
The current rally has survived one major test
The AI-linked technology sector has not moved in a straight line. From 2019 to June 2026, the group gained 569%, but that included a sharp 28.2% drop in 2022. The decline coincided with rising interest rates, tighter financial conditions and a broad reset in long-duration growth stocks.
The rebound that followed was powerful. AI-linked technology shares rose 57.8% in 2023, 36.6% in 2024, 24.0% in 2025 and 19.8% in the first half of 2026. Those gains helped lift major U.S. indexes and reinforced the view that AI had moved from a speculative theme to a central driver of corporate spending.
The broad U.S. equity market also benefited, but not to the same degree. Overall U.S. equities advanced 237% from 2019 to June 2026. That gap shows how much of the market’s leadership has come from companies linked to advanced chips, cloud computing, software, data centers and AI infrastructure.
For traders, the key issue is whether the earnings cycle can continue supporting the price cycle. One strong reporting season may not settle the debate, but a string of disappointing results from chipmakers, cloud providers or large software companies could quickly change sentiment.
Balance sheets separate this cycle from the dot-com era
BlackRock emphasized that corporate balance sheets are one of the biggest differences between today’s AI boom and the internet bubble.
During the dot-com period, many companies funded expansion through heavy borrowing or repeated equity issuance. In some cases, business models were unproven and profits were years away. Capital spending rose far faster than internal cash generation. The ratio of capital expenditure to free cash flow reached roughly four during the internet bubble, meaning companies were spending far more than they were generating.
Today, that ratio remains below one for major technology companies tied to AI. That suggests firms are funding spending on chips, data centers, power systems and cloud infrastructure largely with their own cash flow.
This matters because debt-funded booms can unravel quickly when credit conditions tighten. If companies are forced to refinance, cut spending or sell assets, market weakness can spread through the financial system. A cash-funded boom does not remove valuation risk, but it reduces the chance that unpaid loans alone will trigger a wave of forced retrenchment.
That does not mean AI spending is risk-free. Building data centers, securing electricity, buying advanced semiconductors and training large models are expensive commitments. If demand falls short, even cash-rich companies could face pressure to slow spending. But the starting point is stronger than it was in the late 1990s.
Market leadership remains narrow
Even with stronger balance sheets, concentration remains a serious concern. A small number of large companies continues to carry a large share of index performance. The “Mag 7” group has dominated much of the market conversation in recent years, and the AI boom has created another shorthand: “MANGOS,” referring to Meta, Anthropic, Nvidia, Google, OpenAI and SpaceX.
The comparison is not exact because several of those companies are private, while others are publicly traded through parent companies or directly listed shares. Still, the phrase captures a broader shift in market attention toward firms seen as controlling the most important AI inputs: models, chips, cloud platforms, distribution networks and computing infrastructure.
Morningstar’s global next-generation AI index rose about 45% across April and May 2026 before cooling in June. That move showed how quickly traders continue to chase exposure to AI themes when earnings, product announcements or spending plans appear to support the long-term story.
The risk is that leadership remains too dependent on a narrow set of companies. In 1999, a handful of technology names drove indexes to new highs. When expectations failed, those same stocks dragged the market lower. Today’s leaders are stronger businesses than many dot-com-era firms, but they still face the burden of very high expectations.
AI inputs are becoming a market focus
BlackRock said it continues to prefer assets tied to limited AI inputs, including electricity, semiconductors and data-center infrastructure. That view reflects a practical reality: AI growth requires physical capacity.
The boom is not only about software. It requires advanced chips, cooling systems, high-voltage equipment, fiber networks, land, permitting, water access and enormous amounts of power. Data centers are becoming a major source of electricity demand in several regions, pushing utilities, grid operators and infrastructure companies into the AI discussion.
Semiconductors remain the most visible bottleneck. Advanced graphics processors and custom AI chips are central to training and running large models. When supply is tight, companies that control key chip designs, manufacturing equipment or high-end production capacity can command premium valuations.
Electricity is another constraint. Data centers require reliable power around the clock. That demand has increased interest in utilities, gas generation, nuclear power, renewables, batteries and grid modernization. For traders, this has widened the AI trade beyond the largest technology shares.
Digital assets face indirect pressure from the AI trade
The AI equity rally also matters for cryptocurrency markets, even though the link is indirect. Digital assets often respond to the same global liquidity conditions, risk appetite and technology-sector sentiment that influence high-growth equities.
When traders become more confident in technology earnings, risk appetite can improve across several markets. When concerns rise about valuations, interest rates or earnings disappointments, speculative tokens can come under pressure. That does not mean every digital asset will move with technology shares, but the broad correlation can strengthen during periods of market stress.
Tokens tied to decentralized computing, data storage, cloud resources and physical infrastructure may attract attention because their narratives overlap with demand for AI capacity. Some projects claim to provide distributed processing power, storage networks or data services. However, traders are likely to separate projects with actual usage and revenue from those relying mainly on marketing.
Recent industry data have pointed to growing revenue in some smart digital-token systems during early 2026, with the combined market value of certain infrastructure-linked tokens rising sharply by late April. Still, these markets remain volatile, thinly regulated in many jurisdictions and sensitive to shifts in liquidity.
The key point is that crypto markets should not be treated as insulated from the AI equity cycle. If major technology companies report weaker cloud demand, slower AI monetization or lower margins, that could sour sentiment toward many technology-linked tokens. If earnings remain strong and capital spending continues, infrastructure-linked digital assets may continue to draw speculative interest.
Traders are watching profit reports closely
The next earnings rounds will be important because expectations are already high. Large technology companies are expected to deliver more than 30% earnings growth, according to figures cited by BlackRock. That leaves little room for disappointment.
Markets will be watching for signs that AI revenue is moving beyond heavy spending and into measurable returns. Cloud growth, data-center margins, semiconductor order books, enterprise software adoption and capital-expenditure guidance will all matter.
Traders will also focus on whether spending plans remain internally funded. If free cash flow begins to weaken while capital expenditure keeps climbing, comparisons with the internet bubble could become more uncomfortable. For now, BlackRock argues that major firms still have the cash flow needed to support their AI buildout.
Interest rates are another factor. High valuations are easier to maintain when rates are stable or falling. If borrowing costs rise again, future earnings become less valuable in present terms, placing pressure on growth stocks. Morningstar has already flagged market concentration as a risk, while Fidelity data have supported the view that capital spending remains mostly internally funded.
Bubble fears are not settled
BlackRock’s commentary does not dismiss the dangers surrounding the AI rally. It acknowledges high concentration, stretched long-term valuations and the risk that earnings expectations may become too demanding. But it also argues that the current market is not simply a replay of 2000.
The strongest case for the rally is profit growth. The strongest case against it is valuation. That tension is likely to define markets through the rest of 2026.
For traders, the practical takeaway is that AI has moved from a speculative theme to a central earnings driver for U.S. equities. But the higher the market climbs, the more important each profit report becomes. Strong balance sheets reduce the risk of a debt-driven collapse, yet they do not eliminate the possibility of a sharp valuation reset.
The AI story now needs to keep proving itself in cash flow, margins and productivity gains. If it does, the rally may continue to broaden into chips, power, infrastructure and selected digital-asset sectors tied to real computing demand. If it does not, the same enthusiasm that lifted markets could quickly turn into a source of volatility.
To navigate today’s costly AI‑driven markets like BlackRock describes, explore our in‑depth guide on AI copy trading strategies.
Disclaimer: The content on this page is provided for general informational purposes only and does not represent the views or financial advice of Toobit. We make no guarantees regarding the accuracy or completeness of this information and shall not be held liable for any errors, omissions, or outcomes resulting from its use. Investing in digital assets involves risk; users should independently evaluate their financial situation and the risks involved. For further details, please consult our Terms of Service and Risk Disclosure.
