BRUH "increase InP laser capacity by ~20x from 2025-2030. The vendors appear to have taken a more conservative stance...
Expected multi-year ramp in AI/datacenter optical interconnects should require substantially more InP laser capacity. Public and private reports suggest headline plans of ~20x capacity growth from 2025–2030, but vendors are reportedly taking a more conservative stance (~12x). That mix of strong demand and partial supply discipline supports a constructive but still cyclical setup for laser/component suppliers.
Linked assets
Companies with direct exposure to InP/datacom lasers and optical components could benefit if AI-driven optical networking ramps as expected. COHR offers direct leverage to laser/component demand, while LITE has datacom laser exposure and should participate in a sustained InP capacity/demand upcycle, albeit with cyclicality risk.
Direct leverage to laser/component demand; benefits if AI optical interconnect ramps and supply additions remain rational.
Datacom laser exposure; should participate in any sustained InP capacity/demand upcycle, with cyclicality risk.
Source proof
Source proof: Strong source proof | 3 extracted claims | 2 directional assets | 1 supporting author | headline-like title review
Primary source claims a 2025–2030 InP laser capacity expansion headline of ~20x, with vendors apparently planning a more conservative ~12x build. Related posts provide context on adjacent component markets (MLCC demand for AI servers), PC/laptop unit-growth variability, and other low-information signals. None of the sources provide audited supplier build plans or formal company guidance.
Report: Verizon to provide dark-fiber connectivity for Google data centers in a $1B+ deal; Verizon CEO indicates more similar deals in pipeline. Implication: incremental enterprise/networking revenue and positioning for AI/data-center connectivity demand.
Social post summarizing a SemiAnalysis piece questioning whether AMD can erode NVIDIA’s CUDA moat, citing aggressive pricing ("up to 105% equity rebate" for OpenAI), agentic kernel generation, improving software quality, but also internal cluster instability and MI455X/HelIOS production-ramp risk. Actionable mainly as a sentiment/thesis flag for AMD vs NVDA, but lacks hard numbers/timelines beyond “Advancing AI 2026” and “production ramp hell.”
Social media speculation that “Fable 5.1” will release after “GPT-6,” suggesting a late-August to mid-September timing. No concrete, verifiable product or company details; primarily chatter/banter.
Social post claiming Anthropic Claude “Opus 5” is new state-of-the-art on coding/knowledge work evals and that it is “distilling hard from the Mythos base.” No financials, timelines, product launch details, or adoption indicators are provided.
Social-media discussion reacting to Sam Altman’s statement: he wants the US to “win in AI” across both open-source and proprietary models. No concrete policy, endorsement, funding, or company-specific catalyst is provided, so tradability is limited and mostly reinforces an existing pro-US AI leadership narrative.
Social post claims (unverified) that AMD’s future MI500 platform will use optical interconnects, potentially via Ayar Labs, and could be ahead of Nvidia’s Rubin Ultra in some areas (HBM, 4-die packaging, scale-up). If true, it reinforces a bullish narrative for AMD’s AI accelerator roadmap and for optical-interconnect supply chain, while being modestly bearish for NVDA on relative positioning.
Jensen Huang (NVIDIA) publicly argues that “open models matter,” framing open AI as improving safety/cybersecurity, accelerating innovation/diffusion, and enabling national “sovereignty.” This is narrative-supportive for broad AI buildout (compute demand) and for ecosystems that monetize infrastructure around open models.
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 verifiability on MI500 specifics).
Supporting authors
Single author aggregated multiple social posts and market observations. Supporting items range from quantified market-size/MLCC growth estimates to speculative commentary on model launches and component innovations. Evidence quality varies; the InP-capacity claim is the central actionable datapoint but is sourced from social/posts rather than company filings.
Unlock full thesis monitoring
Consider a buy exposure to select laser/optical-component suppliers with direct InP exposure, sized with awareness of cyclicality and execution risk. Monitor vendor-capacity announcements, OEM optical-module bookings, and short-cycle inventory signals (OEM/distributor builds or destocking) for nearer-term risk updates.