COHR

Post argues that optical transceivers are an underappreciated binding supply constraint inside the AI infrastructure buildout. It claims demand is exceeding supply across the optical stack, that the bottleneck is specialized Indium Phosphide (InP) materials capacity (not easily fixed by capital), and that announced capacity expansions won’t close the gap until late 2027—implying pricing/power upside for constrained optical suppliers and potential deployment-delay risk for AI infrastructure benef

Post argues the AI infrastructure buildout has multiple “floors” of supply-chain constraints. Author claims memory was the key bottleneck in 2025 (more than GPUs/models), cites a large gain in a memory position (“SNDK”), and asserts photonics is the next emerging chokepoint. Actionable mainly as a thematic signal (memory scarcity / photonics constraint), with limited concrete tickers beyond NVDA and the mentioned memory stock symbol.

Post is a concise supply-chain mapping for Co-Packaged Optics (CPO): substrate, epitaxy, laser, optical engine, with a basket of related public tickers (AXTI, LITE, AIXA, COHR, SOI, TSEM). It implies these names are relevant exposures/beneficiaries to a CPO buildout but does not state explicit bullish/bearish positioning, timing, or catalysts.

Post is an educational/market-structure primer on Optical Circuit Switching (OCS) for AI datacenter networks. Core investable implication: OCS is early/mid S-curve with multiple growth vectors (AI growth, adoption, rising attach-rate), which could benefit optical component/switching suppliers explicitly naming Lumentum and Coherent. No explicit valuation, timing catalyst, or trade entry/exit levels; one “hidden smallcap” is referenced but not identified, so not tradable from this text alone.

Post argues AI infrastructure is hitting a connectivity bottleneck (“Copper Wall”) where copper interconnects create heat/latency at scale, making photonics/optical connectivity a critical AI supply-chain. It cites NVIDIA allegedly committing $4B to secure laser capacity from Lumentum and Coherent, and suggests an upcoming Jensen Huang keynote could spotlight photonics as a near-term catalyst. Named public tickers: NVDA, LITE, COHR.

Post is a vague teaser: author dislikes LiDAR companies and suggests they’re entering the NPO/CPO data-center laser market (indium phosphide / DR lasers). No specific catalyst, company names, or timing are provided.

Social posts claim AMD’s next-gen MI500 GPU platform may incorporate optical interconnects and be ahead of Nvidia’s Rubin Ultra in HBM, 4-die packaging, and scale-up domain. Separately, analyst Jeff Pu raises AI accelerator TAM to ~$1.4T by 2030 (from $1T) and lifts 2028 forecast to ~$1T; server CPU TAM >$220B by 2030 with “agentic AI” ~50% of TAM and discussion of CPU:GPU mix. This is high-level/rumor + sell-side TAM framing (directionally bullish for AI compute supply chain, but low verifiabil

Post cites informal channel checks that AMD’s Helios scale-up networking roadmap may adopt co-packaged optics (CPO) in future generations, and that broader hyperscaler/industry adoption of CPO is expected. This is a forward-looking packaging/interconnect thesis that could be bullish for AMD’s AI networking competitiveness and for public optics/photonics suppliers, but lacks specifics (names, timing, sourcing), reducing near-term tradability.

Post argues Google’s earnings imply semiconductor fears were overdone; cites accelerating cloud growth, Gemini MAU jump, and raised 2026 capex guidance as demand/capex signals that should drive “crazy earnings” across select optical/semicap names ($LITE, $COHR) and storage/AI infrastructure-related names ($SNDK, $NBIS). Actionable mainly as a near-term sentiment/capex-readthrough trade from an earnings catalyst, but lacks valuation/entries.

USGS-style overview of germanium: byproduct of zinc processing; demand shifted from early electronics to fiber optics, infrared/night vision, and polymerization catalysts; recent (as of text) consumption > primary production offset by stockpiles and recycling; includes a historical 2000 year-end price ($1,150/kg).

Speaker claims “Physical AI” will be the next supercycle (beyond humanoids) with multi-layered investment opportunities, but provides no specific public tickers tied to Physical AI in the excerpt. Separately, speaker cites an “optical super cycle” as having delivered strong returns and explicitly lists optical-related tickers ($LITE $COHR $CIEN $AAOI $AXTI $SIVE), implying bullish framing and a desire to find the next analogous theme. Most content is thematic and retrospective; only the optical

Source claims InP (indium phosphide) laser manufacturing capacity is planned to rise dramatically from 2025–2030 (headline ~20x), but vendors are reportedly committing to a more conservative ~12x increase. This implies strong expected demand for optical components (AI/datacenter interconnect) while also signaling some supply discipline vs. an aggressive buildout narrative.