Jensen Huang was in Seoul on June 8, meeting South Korean business leaders, when the question about the selloff came. US tech shares had dropped Friday on concerns over a possible interest rate hike. South Korea’s Kospi Index was tumbling that Monday morning as investors continued pulling back from AI bets that had powered one of the strongest bull markets in recent memory.
His answer was not what a cautious executive would give. Huang told reporters the selloff was a buying opportunity, that the AI buildout was only at the beginning, and that investors should be very excited. Within hours, he and SK Hynix announced a multi-year deal to co-develop next-generation AI memory chips, putting action behind the words.
What Jensen Huang said about the AI selloff and what it means for investors
Nvidia CEO Jensen Huang told reporters in Seoul that the global tech selloff should be viewed as an opportunity rather than a warning.
“We’re at the beginning, and whatever happens in the stock market, you should be very happy, because now you can buy at a discount,” he said. “Everyone should be very excited,” he added, according to Bloomberg.
More Nvidia:
- Nvidia is losing an industry that saved it from bankruptcy
- Nvidia CEO makes surprising admission on OpenAI and Anthropic
- Goldman Sachs just found a reason to like Nvidia stock again
The comments came after Huang met with SK Group Chairman Chey Tae-won in Seoul, where he and SK Hynix simultaneously announced a multi-year agreement to co-develop next-generation AI memory chips. Huang framed both the new partnership and the market pullback through the same lens: the AI infrastructure buildout is still at its very beginning, which means current setbacks are noise against a much longer signal, according to Bloomberg.
Why the selloff happened and why Huang sees it differently than the market
The Friday US tech decline that rippled into the Seoul session on Monday was driven by stronger-than-expected jobs data raising the prospect of an interest rate hike. Higher rates compress valuations for growth stocks, particularly AI infrastructure names that have traded at elevated multiples on long-dated earnings expectations. The selloff hit semiconductor stocks hardest because they carry some of the most stretched valuations in the sector.
Huang’s response is rooted in a specific argument about the stage of the AI cycle. He has consistently argued that AI will become foundational infrastructure in the same way electricity and the internet did, and that current capital spending on data centers, chips, and networking represents the early phases of that transition rather than a mature market already pricing in peak returns. From his vantage point, the debate about whether AI stocks are expensive misunderstands where in the cycle the industry actually is, according to Bloomberg.
That argument has been borne out by Nvidia’s own results. The company reported first-quarter revenue of $81 billion, beating analyst estimates and extending a streak of outperformance that has made Nvidia one of the most valuable companies in history.
Nagle/Getty Images
What the Nvidia and SK Hynix memory deal adds to the AI infrastructure picture
The SK Hynix announcement made alongside Huang’s comments is not incidental. Nvidia and SK Hynix confirmed a multi-year collaboration to design future generations of AI memory chips, strengthening SK Hynix’s competitive position against Samsung Electronics in high-bandwidth memory, one of the most constrained components in AI infrastructure, according to CNBC.
High-bandwidth memory is the component that sits closest to the GPU in AI training and inference systems. Demand for it has been so intense that SK Hynix and Micron have been capacity-constrained for months. A multi-year co-development agreement between Nvidia and SK Hynix signals that both companies are planning for AI infrastructure demand that extends well beyond current market conditions, which is precisely the point Huang made in the same breath about why the selloff is a buying opportunity.
Key context on the AI selloff, Nvidia’s position, and the SK Hynix deal:
- The Friday US tech decline that triggered the Seoul session selloff was driven primarily by a stronger-than-expected US jobs report, which raised the probability of a Federal Reserve rate hike; the rate sensitivity of AI stocks reflects their long-duration earnings profiles rather than any fundamental change in demand, according to Bloomberg.
- Nvidia’s Q1 fiscal 2027 revenue of $81 billion beat estimates and marked the company’s fifth consecutive quarter of revenue above Wall Street consensus; the beat-and-raise pattern has continued even as concerns about AI spending sustainability have periodically rattled the stock, according to CNBC.
- SK Hynix supplies roughly 70% of Nvidia’s high-bandwidth memory requirements; the new multi-year co-development agreement deepens a relationship that is already one of the most strategically important supply partnerships in the semiconductor industry, making the announcement alongside Huang’s buyback call more than coincidental, according to CNBC.
- South Korea’s Kospi Index, which Huang was watching fall in real time when he made his comments, has significant exposure to AI semiconductor names including Samsung and SK Hynix; Huang’s decision to make a public buying call in Seoul while the Korean market was actively declining was deliberate, according to Bloomberg.
- Huang’s “we are at the outset of the AI revolution” framing on June 8 is consistent with remarks he made at Computex in May 2026, where he described a future in which every data center, factory, and enterprise would run AI continuously; the consistency across public appearances suggests this is a considered strategic message rather than reactive market commentary, according to Bloomberg.
What Huang’s AI buyback call means for Nvidia stock and the broader market
Huang is not a disinterested observer. He is the CEO of the company most directly tied to the AI infrastructure trade, and his public comments that a selloff is a buying opportunity serve the obvious interest of supporting confidence in his own stock and sector.
Investors should weigh his credibility alongside that context. The track record supports the credibility: every time Huang has made a comparable call about AI demand remaining stronger than markets expect, the subsequent quarters have produced results that backed the assertion.
The more meaningful signal in the June 8 Seoul appearance is the SK Hynix deal. CEOs do not announce multi-year memory chip co-development agreements when they believe demand is peaking. The structure of that deal, combining a public buying call with a long-dated supply partnership, suggests Huang is positioning for a sustained infrastructure cycle rather than managing short-term market psychology.
For investors in Nvidia and related AI infrastructure names, the practical implication of Huang’s June 8 message is straightforward. The selloff was macro-driven, not demand-driven.
If the rate picture stabilizes or the Fed signals patience, the argument for buying the dip in AI infrastructure stocks gains the fundamental support it currently lacks. Whether that happens depends on economic data, not on anything Jensen Huang said in Seoul.
Related: Goldman Sachs delivers clear message to stock market investors