This time, Jensen Huang's speech at GTC actually revealed a very interesting market psychology issue.

Many people have been waiting for a big breakout. The optical communication sector has been hot for a long time, from CPO to silicon photonics, from optical modules to high-speed interconnects—almost all imaginations about AI infrastructure upgrades have been pushed onto this. Coincidentally, OFC 2026 is also happening in the same week—one features NVIDIA's roadmap, and the other showcases the entire optical communication industry chain's muscle. The market's excitement has been driven to a fever pitch. So what everyone was really waiting for wasn't just an ordinary speech, but whether Jensen Huang would say that "optics is the next main trend."

Unfortunately, he didn't.

Of course, Jensen Huang mentioned optics, and he mentioned it quite a bit. But what he really said was that optics is very important, yet copper cables won't be phased out in the short term. NVIDIA plans to continue using copper cables in Vera Rubin Ultra and future systems, while also adopting new optical technologies. Just this small difference was enough to cause a market shift.

This is also the most awkward part of the market. Stocks are often most afraid not of bad news, but of not being as good as expected. The market originally wanted to hear that optics would soon take over completely, but Jensen Huang changed the message to "both optics and copper will be used." This expectation gap directly killed the entire sector's bullish momentum.

But there's a very easy point to misread here: many people interpret this as "optics is no longer viable" or "copper wins." Neither is correct. The long-term logic of optics hasn't changed; what has changed is the market's expectation of how quickly it will be realized. More precisely, Jensen Huang is not correcting the direction but the timeline.

According to his description, copper still dominates in many parts of the rack, but when it comes to larger-scale, cross-rack connections, the importance of optics begins to become more apparent. What does this mean? It's not "optics will take over entirely," but rather "who will be used where first" has become the new question.

After the speech, the market didn't immediately surge across the board but first experienced some turbulence before starting to differentiate. The theme shifted from "as long as it touches optics, it will rise" to "who truly benefits and who is just riding the hype." This differentiation is even more evident at the individual stock level.

Lumentum continues to be discussed by the market not just because it belongs to the optics sector, but because it has already moved beyond being just a concept stock in investors' minds. It has been placed on the list of companies that might enter the next-generation interconnect systems in the future. Even with short-term sentiment fluctuations, the market's understanding remains at the level of "pace change" rather than "logic disappearance."

Coherent is somewhat similar to Lumentum, but its valuation won't be exactly the same. When the sector shifts from "telling big stories" to "focusing on practical implementation," investors will pay more attention to which company benefits at which level, how long it will take to realize, and whether expectations have already been overly optimistic.

Ciena is relatively special. Unlike some highly flexible names that can be quickly pushed up by sentiment and then hammered down, it is more like a stock that prompts the market to think about "how the future optical network will truly unfold." Its significance isn't just riding the trend but reminding everyone that ultimately, the game isn't just about component stories but about how the entire network capability moves forward.

Applied Optoelectronics is a typical high-flexibility representative. When sentiment is good, it’s easiest to push it up quickly, but as long as the catalysts aren't strong enough to continue raising expectations, it’s also the first to face profit-taking pressure. Its volatility precisely illustrates one point: when the market begins to doubt whether realization will be slower than expected, the first to be hit are often those stocks that rose the fastest and had the most optimistic expectations.

Credo exposes another very important change: it’s not just about being connected to copper to automatically benefit. Jensen Huang clarified this time that copper won't exit immediately, but that doesn't mean all copper interconnect companies will be rewarded right away. Going forward, capital will ask more detailed questions: which segment of copper connectivity will benefit the most?

Ultimately, looking at these stocks together, the most noteworthy thing isn't who rises or falls, but that the market has already started to treat them as assets in different positions, with different realization paces and different levels of certainty. Previously, everyone was willing to put them in the same basket, but after GTC, that basket is being unpacked.

AI interconnects are not a "choose between optics and copper" question but a "who uses what where" division of labor. Ultimately, Jensen Huang didn't deny optics; he just didn't say it in the version the market wanted to hear most. So after GTC, the market is no longer just looking at "whether there is a story," but at "who is closer to practical implementation and realization."

The direction of optics hasn't changed; what has changed is how the market views this sector. Previously, people were more willing to buy into the imagination first; now, the focus will shift more toward realization. So, the real gap will be determined not by who is better at storytelling but by who turns the story into actual performance earlier. This is the true differentiation, and it's just beginning.