A relatively quiet day for Serenity, the small-cap stock-picker whose portfolio is built almost entirely around the optical plumbing of the AI buildout. His single note today unpacked a recent podcast appearance ("Market Insights: Photonics & AI") by the chief executive of Sivers (SIVE), the tiny Sweden-listed maker of indium-phosphide continuous-wave lasers that sits at the center of his book. InP is the compound-semiconductor material used for high-speed lasers, and CW lasers are the light source that feeds silicon photonics and co-packaged optics (CPO) — the approach of placing the optical engine right next to the switch or GPU die to cut power and boost bandwidth. Serenity treats Sivers as the overlooked upstream "chokepoint" (a critical supplier with few substitutes) for that transition, and it remains his highest-conviction holding.

The takeaway that stood out to him was reassurance on the durability of pluggable transceivers. There has been a running debate that CPO and various link-optimization techniques would cannibalize the pluggable market, but the Sivers chief reportedly expects pluggables to persist for roughly another decade, with multiple architecture paths — including near-packaged optics (NPO) — coexisting alongside CPO over the next five years. Serenity reads CPO as the industry's eventual destination but pluggables as the larger near-term revenue pool, which matters because it keeps demand broad for qualified laser suppliers.

On the constraint side, the executive pointed to manufacturing capacity and raw-material availability as the main bottleneck — comments Serenity ties directly to the well-documented difficulty of ramping CW laser supply. His conclusion: industry demand is outrunning what all the laser makers can collectively produce, which underpins his view that there is room for every credible supplier.

He also floated a strategic angle. Given the emphasis on the bigger pluggable market, he mused that Sivers acquiring pluggable transceiver IP could make sense — comparing it to how Lumentum (LITE) bolstered its position by buying Cloud Light. Lumentum is the large-cap incumbent of the optical space and a core holding for Serenity, whom he uses as the quality benchmark against which he pitches his higher-upside small caps; it supplies a meaningful slice of Google TPU optics and rides hyperscaler capex.

The note touched two other names in passing. Jabil (JBL), the contract manufacturer whose selection of Sivers lasers for its 1.6-terabit pluggable transceivers is Serenity's key third-party validation of the thesis, was cited as evidence of the pluggable-market focus; he does not own it but views its optical business as underpriced. And Applied Optoelectronics (AAOI) — his Made-in-America, vertically integrated transceiver bet, and a core long — got a caveat: he noted it is also pursuing CPO but sits a bit further behind on commercialization.

Nothing in today's commentary suggests a change in positioning. It reads as thesis maintenance: confirmation that the optical demand curve is steepening faster than supply, that pluggables and CPO will coexist rather than one killing the other, and that the laser-supply chokepoint he has bet on remains real.

*This is derived commentary paraphrasing Serenity's public posts, not investment advice or his own words.*