AI bottleneck research wiki

Reviewable static packet generated from the investor LLM wiki. Current focus: whether glass substrate / TGV / metallization becomes an investable AI physical-stack bottleneck.

Current stance: WATCH / RESEARCHEvidence: mixed primary/vendorConfidence: medium technical, low-medium investableGenerated 2026-05-11 09:54
13
Published synthesis pages
15
Local raw-source references indexed
5
Company/entity pages

Decision supported

Watch / pass / candidate identification for glass substrate, through-glass-via, metallization, bonding, and inspection exposures. No equity graduates to candidate without customer qualification, orders/backlog, revenue materiality, valuation, and downside gates.

Bull case

Glass adoption becomes real for AI/HPC packages; scarce value accrues to hard process steps such as TGV formation, RDL/metallization, bonding, inspection, and qualified fabrication.

Bear case

Evidence remains roadmap/vendor-heavy; TSMC glass linkage is unverified; exposed suppliers may not show material orders, pricing power, or earnings impact.

Next checks

SKC/Absolics filings, LPKF backlog/orders, TSMC symposium/earnings, and customer-side AMD/NVIDIA/Broadcom qualification evidence.

Start here

Published page map

PageTypeConfidenceUpdated
Glass Substrate / TGV / Metallization Deep Divequerymedium2026-05-11
Research Program for AI Bottlenecksquerymedium2026-05-11
AI Bottleneck Betaconceptlow2026-05-11
AI Bottlenecks Dashboardconceptmedium2026-05-11
AI Physical Stack Watchlistconceptlow2026-05-11
Glass Substrate Cycleconceptmedium2026-05-11
Glass Value-chain Maturity Scoreconceptmedium2026-05-11
HBM Manufacturing Bottlenecksconceptlow2026-05-11
Absolics / SKCentitymedium2026-05-11
Intelentitymedium2026-05-11
LPKFentitymedium2026-05-11
SCHOTTentitymedium2026-05-11
TSMCentitymedium2026-05-11

Build fingerprint: 00100e409158

Analytical support only — not individualized financial, legal, or tax advice. Built from local wiki synthesis; raw copyrighted source bodies are not republished.