I've been closely watching the optical communication sector lately. To be honest, the market expectations before GTC were really hyped up. If Jensen Huang can present the optical communication story in a more aggressive way this time, I think the entire sector could surge again. The logic is just too straightforward—AI clusters are growing larger, data center transmission pressures are increasing daily, copper will eventually hit a ceiling, so the next step should be optical, right?

However, Huang's final statement was a bit awkward. He definitely mentioned optical, and emphasized it strongly, but the core message was that copper won't exit the scene in the short term; future systems will use both copper and optical simultaneously. That small difference was enough to cause the market to change its tone. The issue isn't whether optical is problematic; it's that the market initially wanted to hear "optical will take over immediately," but instead heard "copper and optical will coexist." Stocks are most afraid of not necessarily bad news, but of not being as good as expected.

So, after GTC, something interesting happened. The entire optical communication sector is no longer just "rising as long as it gets some attention," but is starting to differentiate genuinely. Companies like Lumentum are no longer seen as just concept stocks; instead, the market believes they have a real chance to enter the next-generation interconnect systems, so even if short-term sentiment fluctuates, the logic remains. Coherent's position is a bit awkward—once the market shifts from "storytelling" to "practical implementation," investors will pay more attention to who truly benefits, when the gains will materialize, and whether the previous expectations were already overly optimistic.

Ciena is somewhat special. Unlike those highly flexible names that can be easily pushed up and then knocked down by sentiment, it’s more like a stock for the market to consider "how will optical networks truly unfold in the future." Applied Optoelectronics is the most typical high-flexibility representative—rising fastest when market enthusiasm is high, but once catalysts aren’t strong enough to push expectations further, it’s also the first to face pullback. The Credo case further illustrates the issue—Huang clarified that copper won't exit immediately, but that doesn’t mean all copper connection companies will be rewarded by the market. 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 with different positions, different payout rhythms, and different levels of certainty. Earlier, everyone was more willing to lump them together, but after GTC, that basket is starting to be unpacked. AI interconnects are not a "choose between optical and copper," but a division of labor based on "who applies where first."

Jensen Huang didn’t deny optical; he just didn’t frame the story in the way the market most wanted to hear. So now, the market is no longer just looking at "whether there’s a story," but at "who is closer to actual implementation." That’s why, even within the optical communication sector, stock performance is beginning to diverge noticeably. The real differentiation has just begun. Previously, investors were more willing to buy based on imagination; now, the market will focus more on actual realization. The true gap in the future won’t be who can tell the best story, but who can turn the story into results earlier.