Is the AI Stock Boom Heading for Another Dotcom-Style Crash?

Wall Street's enthusiasm for artificial intelligence stocks remains remarkably strong despite mounting concerns about profitability, rising borrowing costs and corporate risks.

Wall Street’s enthusiasm for artificial intelligence stocks remains remarkably strong despite mounting concerns about profitability, rising borrowing costs and corporate risks. Yet the history of the dotcom bubble and the 2008 financial crisis offers a warning: overwhelmingly positive analyst ratings do not necessarily mean markets are safe. The current AI boom may be approaching a turning point, but whether it ends in a dramatic crash or a prolonged decline remains uncertain.

Analyst ratings for major AI-related companies show that optimism continues to dominate. On average, 83% of recommendations are currently rated as “Strong Buy”, while only 2% carry a “Sell” rating. Even Nvidia, which lost its CEO Jensen Huang last week following allegations involving multiple extramarital affairs, including one with a suspected Chinese spy, retains strong support from analysts. Around 80% of analysts covering the company rate its shares as a buy, while 20% recommend holding. Only one of 85 analysts has issued a sell recommendation.

For investors, the contrast between these optimistic assessments and the growing number of risks raises an important question: are professional analysts accurately evaluating the AI industry’s prospects, or are they repeating the mistakes made during previous market bubbles?

What the Dotcom Bubble Reveals About Wall Street Optimism

The technology boom of the late 1990s provides one of the clearest examples of how analyst enthusiasm can persist even when valuations become increasingly difficult to justify.

Stay ahead of the geopolitical week.

MD Briefing delivers expert analysis across five global fronts — the Indo-Pacific, energy, geoeconomics, European security, and the Middle East — every Monday morning. Free.

At the beginning of 2000, shortly before the dotcom bubble collapsed, approximately 85% of analyst recommendations for major technology companies were classified as strong buys. These included Amazon, Microsoft, Cisco, Oracle and Vodafone. Sell recommendations were virtually absent.

Some of these companies eventually became highly successful investments. However, their long-term performance does not erase the losses investors experienced during the crash. Several technology stocks suffered substantial declines before recovering, and Cisco did not regain its dotcom-era share price peak until late last year.

The experience demonstrates that identifying a strong company is not the same as identifying the right time to buy its shares. Even businesses with durable technology and long-term growth prospects can become dangerously expensive when investor expectations move too far ahead of financial performance.

The current AI market faces a similar challenge. Companies developing advanced chips, cloud infrastructure and AI applications may benefit from sustained technological demand, but that does not automatically justify every valuation attached to them.

The 2008 Financial Crisis Offers Another Warning

The global financial crisis provides a different example of how analysts can underestimate mounting risks.

At the beginning of 2008, US bank stocks continued to receive relatively favourable ratings, with 53% of analyst recommendations classified as buys and just 12% as sells. This optimism persisted despite several serious warning signs in the housing and mortgage markets.

US house prices had already peaked in 2006. New Century Financial had filed for bankruptcy in April 2007, two Bear Stearns hedge funds collapsed in June because of their exposure to mortgage-related investments, and Countrywide Financial was struggling to survive as mortgage defaults increased.

These developments indicated that the financial system was facing serious structural problems. Nevertheless, analysts remained relatively optimistic about banking stocks.

The episode illustrates a recurring weakness in market forecasting. Analysts may recognise individual risks without fully appreciating how those risks interact or how quickly they can undermine an entire sector.

For the AI industry, the relevant concerns include the cost of building data centres, dependence on continued investment, rising debt financing expenses and uncertainty over how quickly companies can turn AI adoption into sustainable profits.

Will the AI Bubble End With a Crash or a Slow Decline?

Not all financial bubbles collapse in the same way. The dotcom crash and the 2008 financial crisis demonstrate two different paths through which excessive market optimism can unravel.

The dotcom bubble ended with an extraordinary rally followed by a sudden collapse. The Nasdaq surged approximately 95% in the six months between October 1999 and its peak on March 27, 2000. It then lost 32% of its value in just three weeks.

This final surge reflected a powerful wave of investor enthusiasm that pushed technology valuations higher even as concerns about business models and profitability remained unresolved.

The 2008 crisis followed a more gradual pattern. Concerns about the US housing market accumulated over an extended period, gradually weakening confidence in banks and other financial institutions. Between the beginning of 2006 and February 2007, bank stocks advanced by only around 12%. Over the following three months, they moved largely sideways, losing approximately 5% as negative developments continued to emerge.

Eventually, the accumulation of bad news triggered a much broader reassessment of the banking sector and mortgage market. Selling accelerated, turning an extended period of uncertainty into a severe market decline.

These contrasting examples matter because investors often expect bubbles to end with spectacular rallies and dramatic crashes. However, a market can also lose momentum gradually before a sudden shift in confidence accelerates the decline.

What the Recent AI Stock Performance Signals

The performance of the Goldman Sachs US Broad AI basket, which tracks companies associated with the artificial intelligence trade, suggests that the current market has elements of both historical patterns.

Between April and June, the basket surged 51%, driven largely by strong gains in semiconductor stocks. However, it subsequently fell 24% in July and declined another 17% in October.

The sharp movements suggest that investor confidence is becoming more sensitive to negative developments. Concerns about financing costs, the sustainability of AI infrastructure spending and the commercial returns from large technology investments are increasingly influencing market sentiment.

Nvidia’s leadership crisis has added another source of uncertainty. Although analysts continue to maintain overwhelmingly positive ratings for the company, the allegations surrounding Huang have drawn attention to corporate governance and leadership risks at a business central to the AI supply chain.

The recent declines do not necessarily establish that the AI boom has ended. However, they suggest that the market is no longer moving in one direction and that investors are becoming more selective about the risks they are willing to accept.

What Happens Next for AI Investors?

The next phase of the AI market will depend heavily on whether companies can translate enormous investment commitments into sustainable earnings. Continued demand for advanced chips and computing infrastructure may support the industry’s expansion, but the financial burden of maintaining that growth is becoming increasingly important.

Investors will also be watching borrowing costs, corporate earnings, capital expenditure plans and the ability of smaller AI infrastructure companies to secure financing. If revenues fail to keep pace with investment, companies could face pressure to reduce spending, delay projects or seek additional capital under less favourable conditions.

The recent decline in AI stocks could prove to be a temporary correction rather than the beginning of a broader collapse. Equally, it could represent an early stage in a longer reassessment of valuations.

The historical lesson is that strong analyst recommendations offer little protection when market expectations become disconnected from financial realities. The dotcom bubble demonstrated how quickly enthusiasm can turn into panic, while the 2008 crisis showed how prolonged warning signs can eventually produce a sudden reversal.

For now, the AI boom remains supported by genuine technological developments and substantial investor interest. But whether it ends with a dramatic crash or a gradual loss of momentum may depend on how effectively the industry converts its extraordinary promises into measurable economic returns.

With information from Reuters.

Sana Khan
Sana Khan
Sana Khan is the News Editor at Modern Diplomacy. She is a political analyst and researcher focusing on global security, foreign policy, and power politics, driven by a passion for evidence-based analysis. Her work explores how strategic and technological shifts shape the international order.