AI optics: bandwidth becomes supplier earnings · Campaigns
AI optics: bandwidth becomes supplier earnings
active · AI · Optics
Reviews
The case as it stood at each review, with the original call references and current outcomes.
Scope review: AI optics: bandwidth becomes supplier earnings
Campaign review · 2026-09-06 · data as of 2026-09-04
Promote to active research using current supplier evidence; keep the commercial optical case separate from power and memory.
Promote to active research using current supplier evidence; keep the commercial optical case separate from power and memory. Scope reviewed on September 6; source dates remain original. This is a research-priority and evidence-boundary review, not an entry call or a fresh historical performance grade.
0 original calls referenced · 0/0 directional calls have 30-day outcomes · 0/0 directional calls have 90-day outcomes
Case at review: Increasing optical content in AI networks will sustain supplier demand and margin gains through the next two reporting cycles, before added capacity or substitution removes the advantage.
Lumentum and Coherent provide current issuer evidence of demand and product progress; Coherent also reports expanding capacity, making supply response an immediate test.
Horizon: 6–18 months; the shared test with the AI-spending counter-case runs through March 2027.
Failure conditions: Capacity relief or substitution causes order weakness and margin deterioration across two supplier reports · By March 2027, spending cuts reverse demand and margin gains at both core optical suppliers
- Earlier on a material supplier allocation, product-ramp, cancellation or capacity update; separate management guidance from delivered results.
Optical supercycle review: hold; no supplier print since the baseline, Thailand report unconfirmed
Campaign review · 2026-09-30 · data as of 2026-09-30
Due 09-30 review; supplier news read, hold
Hold. Fresh news pulls for COHR, LITE, CIEN, FN and CRDO (fetched 2026-10-01 UTC, 09-30 PT) contain mostly quote pages and nothing material dated after 09-06. The filing watch (09-30) shows only Form 4s from the core names, no 8-Ks. Neither failure test has started. The last instrument prints predate the 09-06 baseline. Ciena (09-03) beat EPS consensus ($2.11 vs $1.73 per yfinance, basis not stated) with record backlog and supply-limited guidance. Credo (09-01) beat on revenue ($479M) but fell about 19% on press-reported margin and guidance worries, which no source attributes to capacity relief or substitution. That is a watch item, not a failure leg. Coherent's PhotonLink launch (ECOC, 09-21) is a product debut with no ramp yet. Open risk: TheBuildout (09-27) reported Thai flood disruption at Fabrinet, Lumentum and Coherent. It rests on a tweet about hard-drive plants, has no issuer confirmation, and is unchecked, not refuted. Price weakness on the 09-30 settled tape (FN 20.2% and CIEN 9.7% below their 200-day averages; COHR at its 200-day) is context, not supplier evidence. LITE's trend stays intact (+496.93% over 1Y, desk summaries).
0 original calls referenced · 0/0 directional calls have 30-day outcomes · 0/0 directional calls have 90-day outcomes
Case at review: Increasing optical content in AI networks will sustain supplier demand and margin gains through the next two reporting cycles, before added capacity or substitution removes the advantage.
Lumentum and Coherent provide current issuer evidence of demand and product progress; Coherent also reports expanding capacity, making supply response an immediate test.
Horizon: 6–18 months; the shared test with the AI-spending counter-case runs through March 2027.
Failure conditions: Capacity relief or substitution causes order weakness and margin deterioration across two supplier reports · By March 2027, spending cuts reverse demand and margin gains at both core optical suppliers
- Confirm or deny Thai flood impact on FN, LITE and COHR from issuer releases or 8-Ks
- FN print 2026-11-02 (provider estimate): orders, data-center revenue, margin, capacity
- COHR 11-04 (estimate) and LITE 11-05 (calendar date): allocation, capacity response, margin; keep guidance separate from delivered
- CRDO 11-30: does the margin concern persist, and is it tied to substitution?
- Optional read-through: GLW 10-27 (non-instrument)
Optical supercycle review: hold; Lumentum's sold-out claim is secondary guidance, not delivered results
Campaign review · 2026-10-16 · data as of 2026-10-16
EOD 2026-10-16 checkpoint: capacity-update clause fired on secondary guidance (held; linked as context)
Hold, with the new items linked as context only. The early capacity clause fired on secondary reports. Lumentum's CEO told Bloomberg its optical capacity is effectively sold out through early 2029; the remark was relayed on X on October 8-9 and repeated by an INDmoney quote page and TheBuildout's AI-written gloss. An October 2 X transcription of his Global Photonics Economic Forum remarks has 2027 supply covering about 30% of demand. No Lumentum release, filing or transcript carries either statement, and the October 16 filing watch shows no 8-K from Lumentum, Coherent, Fabrinet or Ciena in the prior 14 days. These are management guidance about allocation, not delivered results. They fit the demand-absorbs-capacity leg but establish nothing about orders, delivered revenue, margin or price terms. Convequity (October 9) argues Lumentum's laser demand depends on co-packaged-optics designs that are still open; that names the substitution channel in the first failure test but carries no order data. Neither failure leg has started. The two-report test opens with Fabrinet's report, provider-estimated for November 2, and needs Coherent (November 4, after the close, per the company's investor-relations page) or Lumentum (November 5, provider calendar) to complete. The October 16 settled tape is a cohort rout, not supplier evidence: the optical list was 0 of 28 up on the week and on Friday. Lumentum is a deep pullback in an intact trend (+404.89% over one year, 6.0% above its 200-day average). Coherent (+100.85% one year, 20.9% under its 200-day), Ciena (+94.83%, 16.8% under) and Credo (+24.3%, 5.8% under) are long trends under test. Fabrinet is the broken one on the long horizon (-11.87% one year, trailing one-year alpha -27.4, 31.5% under its 200-day) going into the report that opens the test; no issuer statement on file explains the drop. The Thai flood report still has no issuer confirmation. Credo's October 16 8-K has not been read.
0 original calls referenced · 0/0 directional calls have 30-day outcomes · 0/0 directional calls have 90-day outcomes
Case at review: Increasing optical content in AI networks will sustain supplier demand and margin gains through the next two reporting cycles, before added capacity or substitution removes the advantage.
Lumentum and Coherent provide current issuer evidence of demand and product progress; Coherent also reports expanding capacity, making supply response an immediate test.
Horizon: 6–18 months; the shared test with the AI-spending counter-case runs through March 2027.
Failure conditions: Capacity relief or substitution causes order weakness and margin deterioration across two supplier reports · By March 2027, spending cuts reverse demand and margin gains at both core optical suppliers
- Find a primary source for Lumentum's capacity statements (the Bloomberg interview of October 8-9, the Global Photonics Economic Forum recording, or the November 5 release and call); until then treat 'sold out through early 2029' as secondary guidance
- FN 2026-11-02 (provider estimate, observed 2026-08-18): orders, data-center revenue, margin, capacity; opens the two-report test
- COHR 11-04 after the close (company investor-relations page) and LITE 11-05 (provider calendar): delivered optical revenue and margin against guidance, pricing, capacity adds, any co-packaged-optics or substitution comment
- Read Credo's 8-K filed 2026-10-16 and classify it (allocation, ramp, cancellation, capacity, or none)
- CRDO 11-30 (provider calendar): does the margin concern persist, and is it tied to substitution
- Thai flood: no issuer confirmation as of 10-16; keep the early trigger