Shares of SK Hynix plunged more than 15% in Seoul on Monday, marking the company’s worst single-day decline on record, as investors rushed to lock in profits following its blockbuster Nasdaq debut and reassessed valuations amid uncertainty over the next phase of the artificial intelligence investment cycle.

The memory-chip maker’s shares closed 15.4% lower, their steepest one-day decline since listing, according to LSEG data. The sell-off came just one trading session after SK Hynix’s American depositary receipts (ADRs) surged 13% in their Wall Street debut, underscoring the sharp divergence in investor sentiment between U.S. and South Korean markets.

Rather than signaling a deterioration in the company’s business fundamentals, analysts said the decline reflected a combination of profit-taking, valuation adjustments, and the mechanics of one of the largest equity offerings in history.

Investors Reassess Valuations After Landmark U.S. Listing

SK Hynix’s record $26.5 billion Nasdaq listing has given global investors a new benchmark for valuing the world’s leading producer of high-bandwidth memory (HBM) chips, a critical component used in AI accelerators designed by customers including Nvidia.

The U.S. listing has also created an unusual pricing gap between the company’s Seoul-listed shares and its newly traded ADRs.

Daniel Yoo, Global Strategist at Yuanta Securities, said investors are still trying to determine how the two listings should trade relative to each other.

“Everybody’s really confused about what’s going to happen to the memory demand and where the fair price is,” he said, adding that the market remains focused on the industry’s central question.

“It’s all about how much demand is there versus how much supply is going to come in … [and] what kind of multiple you will be getting.”

There has been growing uncertainty over the sustainability of the AI-driven semiconductor rally after one of the strongest runs ever recorded in memory-chip stocks.

According to Yoo, investors are also comparing SK Hynix’s valuation with that of other global semiconductor leaders.

He noted that Taiwan Semiconductor Manufacturing Company’s U.S.-listed ADRs typically trade at a 13% to 14% premium to their Taiwan-listed shares.

By contrast, SK Hynix’s debut has left a valuation gap exceeding 20% between its U.S. ADRs and domestic shares, prompting investors to reposition holdings as they attempt to determine where the two securities should ultimately converge.

The ADR listing effectively introduces a second reference price for the company, encouraging arbitrage activity and increasing short-term volatility until investors settle on an appropriate valuation relationship.

New Share Issuance Weighed On Domestic Stock

Analysts also pointed to the structure of the offering itself. Unlike a secondary sale of existing shares, SK Hynix issued new stock through the ADR transaction, increasing the number of shares available to investors.

Yoo said the additional equity supply contributed to selling pressure in Seoul.

“The market is taking this as a correctional period for SK Hynix domestically,” he said.

Such price adjustments are common following large equity offerings as markets absorb the increase in outstanding shares and incorporate any resulting dilution into valuations.

Despite Monday’s sharp decline, analysts largely maintained a constructive outlook for the company, arguing that the long-term investment case remains supported by robust demand for AI memory.

SK Hynix has emerged as one of the biggest beneficiaries of the global AI boom because it dominates the market for high-bandwidth memory chips used in advanced AI processors.

Demand for HBM continues to exceed industry supply as hyperscale cloud providers and technology companies accelerate spending on AI infrastructure.

Yoo expects that imbalance to continue supporting earnings over the medium term.

He said the recent weakness is likely to prove temporary, adding that the stock should move “in the right direction” over the next six to twelve months as structural demand continues to outpace new production capacity.

Phillip Wool, Chief Research Officer at Rayliant Global Advisors, also argued that the sell-off should not be interpreted as waning confidence in artificial intelligence.

Instead, he said investors were simply reducing oversized positions after extraordinary gains across the semiconductor sector.

“I think it’s mostly risk management,” he said.

According to Wool, many institutional investors had become heavily concentrated in South Korean and Taiwanese semiconductor stocks following months of exceptional performance.

“Prudent risk management suggests you have to scale those back,” he said, adding that the recent selling “doesn’t really speak to any sort of reduction in the excitement about AI hardware.”

Rather, the AI investment theme is expanding beyond semiconductor manufacturers into software, networking infrastructure, power systems and enterprise AI applications.

Although investors are increasingly diversifying across the AI value chain, memory manufacturers remain among the industry’s most strategically important companies.

High-bandwidth memory has become a critical bottleneck for AI hardware because it enables the rapid movement of enormous volumes of data between processors during model training and inference.

The persistent shortage of HBM has allowed suppliers such as SK Hynix to command premium pricing and expand margins as technology giants continue investing aggressively in next-generation AI infrastructure.

While questions remain over how quickly competitors can add production capacity, analysts broadly expect demand for advanced memory to remain strong as companies race to build larger data centers and deploy increasingly sophisticated AI models.

Monday’s record decline, therefore, appears less a reflection of weakening industry fundamentals than a technical correction following an exceptional rally. After soaring on the back of one of the largest foreign listings in U.S. history and extraordinary investor enthusiasm for AI, SK Hynix’s domestic shares are undergoing a period of price discovery as markets reconcile valuations between Seoul and New York.

With AI infrastructure spending continuing at unprecedented levels and memory supply expected to remain constrained, many analysts believe the company’s long-term growth outlook remains intact, even if heightened volatility persists in the near term.

Investor Michael Burry has launched a sharp attack on prediction markets such as Kalshi, saying that they are effectively gambling platforms operating through a regulatory loophole and warning that retail users are vulnerable to unfair trading practices.

Burry speaks as prediction markets face growing scrutiny from researchers, lawmakers and investors over whether the platforms provide a level playing field for ordinary users and whether they should be regulated more like traditional sports betting or financial markets.

The hedge fund manager, best known for predicting the 2008 U.S. housing market collapse and whose story was chronicled in The Big Short, shared his views in a series of posts on X, shifting his focus from financial markets to the rapidly expanding prediction market industry.

“Kalshi as with all prediction markets is in a regulatory loophole within an extremely heavily taxed and excessively regulated industry that is gambling no matter what anyone calls it,” Burry wrote.

Questions Over Regulation

Prediction markets allow users to buy and sell contracts tied to the outcome of future events, ranging from elections and economic data to sporting events and weather forecasts. Unlike conventional sportsbooks, platforms such as Kalshi operate under the oversight of the U.S. Commodity Futures Trading Commission (CFTC) because they structure their offerings as event contracts rather than traditional wagers.

That regulatory distinction has enabled prediction markets to expand rapidly across the United States, even as sports betting remains subject to state-by-state regulation.

Critics argue that, regardless of their legal structure, many event contracts function much like conventional gambling products. Burry echoed that criticism, saying the current framework allows prediction markets to operate outside many of the restrictions imposed on traditional gambling operators.

Burry also challenged claims by Kalshi that its markets are better equipped to detect insider trading than traditional financial markets.

Kalshi Chief Executive Tarek Mansour has argued that suspicious trading activity is easier to identify on prediction platforms because of the transparency of their markets and transaction data.

Burry rejected that argument.

“Kalshi presents the ability to gamble and cheat, and the loophole allows all of it in every state,” he wrote.

He added that the industry’s rapid growth reflects regulatory gaps rather than superior market design.

“Of course it is number 1 by a mile and will keep growing as long as society steps aside and lets these horrific base human weaknesses run roughshod over all and any logical, moral and decent challenge.”

In another post, Burry argued that prediction markets lack sufficient safeguards against manipulation.

“There is nothing preventing cheating in prediction markets. Cheating, the only activity as old as gambling, with the same power to drive humans to extremes.”

Burry’s criticism follows renewed debate over whether retail participants can compete fairly against professional traders on prediction platforms.

Last week, the Roosevelt Institute published research estimating that ordinary users have collectively lost nearly $600 million on Kalshi since 2018. The think tank noted that a relatively small group of sophisticated traders has captured most of the profits, leaving casual participants at a structural disadvantage.

Kalshi has disputed aspects of the criticism, maintaining that its markets are transparent and operate under federal oversight.

The debate bolsters one of the industry’s central questions: whether prediction markets primarily serve as information-discovery mechanisms or whether they increasingly resemble speculative gambling platforms where experienced traders consistently outperform retail participants.

Burry is not alone in raising concerns about the industry’s regulatory framework.

Distressed debt investor Thomas Braziel has also argued that prediction markets face significant regulatory risks, describing their business model as a form of regulatory arbitrage designed to avoid state gambling laws. He has warned that future regulatory changes could materially affect the industry’s growth.

The criticism comes as prediction markets continue to expand rapidly.

Trading volumes on leading platforms, including Kalshi and Polymarket, have surged over the past year as users increasingly speculate on elections, economic releases, geopolitical developments and sporting events. The sector has attracted growing interest from retail traders seeking alternatives to traditional investing and sports betting.

At the same time, lawmakers have introduced several proposals aimed at tightening oversight of event contracts, particularly those tied to sports and elections. Some bills would prohibit certain categories of prediction contracts altogether, while others seek to establish clearer regulatory boundaries between financial products and gambling.