GLW | Earnings Review — Q2 2026

Verdict: ACCELERATING on the metrics that matter — Enterprise Gen-AI optical and EPS leverage — with core sales stabilized in a high-teens plateau. Core sales of $4,738M grew +17.1% YoY and core EPS of $0.78 grew +30.0% YoY. Optical Communications hit $2,072M (+32.3%), led by Enterprise optical at $1,269M (+65.0%). Core gross margin expanded +120 bps to 39.6%; core operating margin +190 bps to 20.9%. Management called results “outstanding,” still scale-out only — scale-up and photonics are not yet in the print.

Double beat, surprise magnitude re-widening. Core sales beat consensus by +$108M / +2.3% and cleared the ~$4.6B guide. Core EPS beat by +$0.025 / +3.3% and printed above the high end of the $0.73–$0.77 guide despite a known ~$30M Solar wafer-facility maintenance drag. This is the 10th consecutive core EPS beat since Q1 2024; after three thinner surprises (+0.8% / +1.8% / +1.2%), Q2 re-accelerated magnitude to +3.3%.

New guidance keeps the Springboard path intact — street is parked at the EPS floor. Q3 core sales $4.9B–$5.0B (~+16% YoY); core EPS $0.85–$0.89 (~+28% YoY). FMP street sales $4.983B sits near the high end of the range; street EPS $0.85 sits at the low end — a historically favorable setup for Corning’s beat streak. Guide midpoint annualizes near the $20B end-2026 run-rate a quarter early. FY CapEx raised to ~$2.0B from ~$1.7B to fund optical capacity under customer LTAs.

Tone. From Q1’s “excellent / Springboard upgrade coming” to Q2’s “outstanding / delivering against 20-30-40.” Confidence is highest on Enterprise orders and Q3 continuity; deliberately conservative on formal multi-year OM re-target (“later this year”) and on near-term Photonics dollars (no $1B-next-year company guide).

Contradictions (5). One high: Q4’25 Display “exceeded 25% NI margin” while stating 17% in the same sentence (quarterly math implies ~26.9% — likely ASR/transcript integrity). Two medium live narrative tensions: “accelerating growth / orders” vs Q3 sales guide that is ~100 bps slower YoY (+17% → ~+16%); Carrier “accelerating / nice growth” language vs +1% YoY Q2 print (H1 +17% reconciles the math). CapEx step-up supersedes original “capacity already in place” Springboard pitch. One low ASR slip ($30M/$40M vs $30B/$40B).

Near-term catalysts. Q3 print (2026-10-27) on Enterprise sustain + Solar profit turn; $20B run-rate timing; capacity/LTA milestones (Meta ≤$6B, NVIDIA US optical mfg, Amazon multi-billion DC fiber); Photonics/Scale-Up remains 2027+ optionality, not in results.

Consolidated core metrics

What is accelerating vs decelerating

| Series | Path | Read | |---|---|---| | Core sales YoY | Mid-teens trough (H1'25) → +17–18% H1'26 | Re-accelerated off 2025 mid-cycle dip; Q2 −97 bps vs Q1 is plateau noise, not a break | | Optical YoY | +24% trough (Q4'25) → +32–36% H1'26 | Re-accelerating; Enterprise re-kicked to +65% | | Carrier optical | +36% (Q1'26) → +0.8% (Q2'26) | Decelerating on project timing; H1 still +17% | | Core OM YoY bps | +130 to +250 bps for 8 straight quarters | Sustained expansion | | Core EPS YoY | Mid-20s trough → +30% Q2'26 | Re-accelerating three straight; still ~1.8× sales growth |

YoY growth chart — core sales, Optical, core EPS

Annual frame (FY2021–FY2025)

| Metric | FY2021 | FY2022 | FY2023 | FY2024 | FY2025 | |---|---:|---:|---:|---:|---:| | Core sales ($M) | $14,120 | $14,805 | $13,580 | $14,469 | $16,408 | | Core sales YoY | — | +4.9% | −8.3% | +6.5% | +13.4% | | Optical sales ($M) | $4,349 | $5,023 | $4,012 | $4,657 | $6,274 | | Optical YoY | — | +15.5% | −20.1% | +16.1% | +34.7% | | Core OM % | 17.4% | 16.8% | 16.5% | 17.5% | 19.3% | | Core EPS | $2.07 | $2.09 | $1.70 | $1.96 | $2.52 | | Core EPS YoY | — | +1.0% | −18.7% | +15.3% | +28.6% |

FY23 was the trough; FY25 shows the recovery (core sales +13.4%, Optical +34.7%, EPS +28.6%). Run-rate into 2026 is faster than FY25 full-year growth.

This quarter vs consensus and guide

| Metric | Consensus | Actual | Variance | vs company guide | |---|---:|---:|---:|---| | Core sales | $4.630B | $4.738B | +$108M / +2.3% BEAT | Above ~$4.6B | | Core EPS | $0.755 | $0.78 | +$0.025 / +3.3% BEAT | Above high of $0.73–$0.77 |

Management’s variance explanation: Optical / Enterprise Gen-AI demand and margin leverage more than absorbed Solar maintenance (~$30M incremental vs Q1), soft Glass (+1%), and muted Auto (+2%). Still scale-out only.

EPS beat heatmap — last 8 quarters (this quarter highlighted)

| Window | EPS beats | Rate | Notes | |---|---:|---:|---| | L4Q (Q3'25–Q2'26) | 4 / 4 | 100% | Q2'26 largest of the four | | Streak Q1'24→Q2'26 | 10 / 10 | 100% | Consistent beater | | L4Q avg surprise | +1.8% | — | Q2'26 +3.3% re-widens after thin prints |

Pattern: Consistent beater. Magnitude compressed vs H1'25 peaks (+5–6.5%) then re-accelerated this quarter — improving, not deteriorating. Revenue flipped to three straight core sales beats (Q4'25–Q2'26).

Q3 2026 company guide (issued 2026-07-28)

| Item | Guide | Implied YoY | Street (FMP) | vs street | |---|---|---|---|---| | Core sales | $4.9B–$5.0B (mid $4.95B) | ~+16% | $4.983B | Mid ~flat / −0.7% vs street | | Core EPS | $0.85–$0.89 (mid $0.87) | ~+28% | $0.85 | Mid +2.4% above street; street = guide low | | FY CapEx | ~$2.0B | Raised from ~$1.7B | — | Optical capacity under LTAs | | Tax rate (approx.) | 18.5% | −50 bps vs prior 19.0% | — | Mild EPS tailwind | | $20B run-rate | $20B | Possibly Q3 early | Street FY sales $19.24B (reported) | Narrative / multiple catalyst |

Waterfall — levels ($B sales / $ EPS)

Bridge map

| Step | Sales | EPS | Driver | |---|---|---|---| | Prior Q2 guide → Q2 actual | +$0.14B | +$0.03 vs mid | Enterprise optical / Gen AI; solar drag absorbed | | Q2 actual → Q3 guide mid | +$0.21B | +$0.09 | Volume + Solar profit recovery + optical mix | | Q3 mid vs consensus | −$0.03B | +$0.02 | Sales ~in-line; EPS guide above street | | CapEx prior → new FY | — | — | $1.7B → $2.0B optical growth invest |

Tone vs Q1: more assertive execution confidence; risk set rotated (Solar de-risked; memory→handheld mid-teens unit headwind and photonics timing elevated). Management explicitly rejects reading Q3 as a deceleration signal.

Other forward statements: Springboard internal 20-30-40 reaffirmed; high-confidence investor plan $27B (’28) / $35B (’30) unchanged; phase-2 sales CAGR 19% (Q4’26→Q4’30); Photonics $10B MAP by 2030, no FY27 Photonics $; Solar sales and profit improve from Q3; OM formal re-target deferred to later 2026.

8-quarter core growth trajectory

| Metric | Q3'24 | Q4'24 | Q1'25 | Q2'25 | Q3'25 | Q4'25 | Q1'26 | Q2'26 | |---|---:|---:|---:|---:|---:|---:|---:|---:| | Revenue YoY % | +7.9% | +18.4% | +12.9% | +12.2% | +14.4% | +13.9% | +18.1% | +17.1% | | Rev Accel (bps QoQ) | +442 | +1,048 | −548 | −68 | +220 | −55 | +421 | −97 | | EPS YoY % | +20.0% | +46.2% | +42.1% | +27.7% | +24.1% | +26.3% | +29.6% | +30.0% | | EPS Accel (bps QoQ) | +1,556 | +2,615 | −404 | −1,445 | −359 | +225 | +331 | +37 |

Accel = change in the YoY rate vs the prior quarter (bps), not sequential sales growth.

| Inflection | What happened | Signal | |---|---|---| | Q4’24 | Sales YoY jumps to +18.4%; EPS peaks +46% | Early Springboard re-acceleration | | Q1–Q2’25 | Sales cools to ~12–13%; EPS rolls hard | Base effect; still double-digit sales | | Q4’25 → Q1’26 | Sales re-accel to +18.1%; EPS turns up | Second inflection — durable mid/high-teens + EPS reaccel | | Q2’26 | Sales +17.1% (−97 bps); EPS +30% (+37 bps) | Stabilizes elevated; EPS still grinding higher |

Net trajectory call: Positive and resilient. Sales hold a high-quality high-teens band while EPS has reaccelerated to ~30% for three straight quarters. Springboard thesis (earnings grow faster than sales) is visible in the data. Q3 guide (~+16% sales / ~+28% EPS) keeps the path intact.

Key drivers this print: Optical engine (Enterprise Gen AI); hyperscale LTAs (Meta / NVIDIA / Amazon / Apple); Solar temporary P&L drag with Q3 recovery expected; Glass/Auto low-single-digit ballast; margin expansion and ROIC +180 bps to 14.9%.

| # | Catalyst | Timing | Consensus / market view | Mgmt signal | Watch | |---|---|---|---|---|---| | 1 | Enterprise Gen-AI scale-out | Q3–Q4'26 | Primary growth engine; sequential durability questioned | Orders accelerating; Gen-AI product sales nearly doubled; still scale-out only | Can Enterprise $ keep climbing without air-pocket? | | 2 | Solar profit turn | Q3'26+ | Street EPS at guide low may embed residual drag | Sales + profit improve from Q3 after ~$30M Q2 maintenance | Cleanest discrete Q3 P&L swing toward $0.89 | | 3 | $20B annualized run-rate | Possibly Q3'26 | Street FY $19.24B is reported, not run-rate | Guide mid ~$4.95B → ~$19.8B annualized; may hit early | Sentiment / multiple catalyst | | 4 | Optical capacity + margins | H2'26 CapEx step-up | Wants Optical NPAT leverage to continue (Q2 NI +77% YoY) | Optical NPAT margin 21% record; CapEx ~$2.0B | Q3 GM/OM step-up (Solar + mix) | | 5 | Scale-Up / Photonics MAP | 2027+ material $ | Street models ~$1B+ “next year” scenarios; timing chatter | No change vs May; no company FY Photonics $; $10B by 2030 | Largest long-duration option; not in Q3 P&L | | 6 | Customer LTAs | Multi-year | Proof points for Springboard; CapEx co-funded | Meta ≤$6B; NVIDIA US mfg; Amazon multi-billion DC; Apple 100% cover glass KY | Capacity commissioning + more LTAs | | 7 | Carrier re-acceleration | H2'26 / 2027 | Q2 soft print raised reallocation questions | +1% Q2 but H1 +17%; MSD longer-term intact | Project timing, not structural | | 8 | OM target reset | Later 2026 | Still anchors to ≥20%; Q2 already 20.9% | Will return later after more Solar / Photonics history | Upside IR catalyst if formal bar lifts | | 9 | Q3 earnings print | 2026-10-27 | Sales $4.983B / EPS $0.85 | Guide $4.9–5.0B / $0.85–0.89; “no deceleration” | Highest-probability calendar catalyst | | 10 | Citi TMT | 2026-09-09 | Soft color event | Confirmed attendance | Photonics timing language |

Ranked watch-list: (1) Q3 print on Enterprise + Solar + Optical NPAT; (2) capacity / LTA milestones; (3) Scale-Up architecture news in the NVIDIA ecosystem; (4) Sept TMT color; (5) year-end Springboard / margin update.

Scorecard: 7 analysts / 12 sub-questions — ~67% well answered, ~33% deflected. Tone confident, unhurried; repeatedly killed “implied deceleration.”

Well answered (highlights)

| Analyst | Ask | Answer quality | |---|---|---| | Asiya Merchant (Citi) | Why limited Q4 growth after ~$5B exit? Hidden soft markets? | Well answered — $20B possibly a quarter early; not re-upgrading Springboard two months after May; no weakness in other MAPs | | Asiya (Citi) | Optical NPAT margins H2 | Well answered — 20%→21%; further expansion expected | | Josh Spector (UBS) | Photonics/scale-up vs May; $1B+ next year? | Well answered — no change vs May; explicit correction: company did not guide $1B Photonics next year; long-term $10B intact | | Wamsi Mohan (BofA) | Q3 deceleration implied? | Well answered — guide not intended to imply deceleration; Enterprise remains significant in Q3 | | Joe Cardoso (JPM) | Q3 margin step-up fair? | Well answered — yes; Solar named as discrete Q2→Q3 driver; ≥20% OM; new target later | | Meta Marshall (MS) | Carrier +1% vs MSD longer-term | Well answered — H1 Carrier +17% / +$0.25B; project timing; FTTH + DCI demand growing | | Mehdi Hosseini (Susq.) | More poly to e-grade semi? | Well answered — will increase highest-grade semi poly; solar volume still dominates $ |

Deflected / avoided (4)

| Analyst | Unanswered core ask | What they got instead | |---|---|---| | George Notter (Wolfe) | % of Optical under LTA today / in 12 months | “Lion’s share” trajectory; capacity LTA-backed | | Wamsi Mohan (BofA) | Why ~$300M QoQ Enterprise didn’t drop through to leverage in Q2 | Forward Optical margin optimism (no Q2 bridge) | | Meta Marshall (MS) | Verizon / Google named project participation | Generic product-set / franchise language | | Mehdi Hosseini (Susq.) | How much Springboard is contingent on scale-up / FAU | Architecture taxonomy; no % bake-in of 20-30-40 |

Pattern: Transparent on direction (growth, margins, LTAs rising, photonics long-term) and systematically non-transparent on levels that let the street stress-test Springboard (LTA %, scale-up %, intermediate Photonics $). Highest-stakes bull case only partially verifiable from the call.

Five items ranked most-severe first. Red-border alert cards below.

Cleared non-contradictions: dual-track Springboard (internal 20-30-40 vs high-confidence $27B/$35B); scale-up discussed but not in results; Carrier Q1 +36% vs Q2 +1% (different quarters; H1 +17%); Solar multi-year target vs temporary Q2 loss.

Macro — two-speed economy

| Theme | Management signal (Q2'26) | Read-through | |---|---|---| | AI / data-center infrastructure | Enterprise +65%; AI portion of enterprise nearly doubled; orders accelerating; “if we could make more, we could sell more” | Very strong — demand still capacity-tight for high-density optical through at least H2'26 | | Consumer / handheld | Memory prices → handheld units mid-teens % down FY; Gorilla outperforms via content | Soft unit CE cycle; premium content partially offsets | | Display / TV | Flat units assumed; brands shifting to larger / higher-price TVs (Gen 10.5 favored) | Soft units, premium mix | | Auto | Global vehicle market −2%; Corning auto +2%; Class-8 diesel +3% YoY / +13% seq | Light-vehicle muted; NA heavy-truck orders improving | | Solar / industrial policy | Preference for U.S.-made solar strengthening (trade/tax) | Policy tailwind for domestic poly/wafer | | Semi / reshoring | More poly into highest-grade semi over time; advanced memory demand supports optics | Onshoring + advanced-memory support specialty inputs | | Rates / Fed | Not addressed | No monetary-policy framing this print |

Named third parties

| Company | Relationship | Implication | |---|---|---| | Apple (AAPL) | 100% iPhone/Watch cover glass at KY | Deepens domestic content stickiness even if industry units fall | | Meta (META) | Multiyear optical up to $6B | Bullish on Meta AI infra spend durability | | NVIDIA (NVDA) | Multiyear US optical mfg partnership; Vera Rubin Ultra hybrid scale-up roadmap | Validates optical scale-up path; still scale-out only in GLW results today (up to ~10× fiber/GPU in full optical scale-up case) | | Amazon (AMZN) | Multibillion US DC optical supply | Hyperscale fiber intensity converting to multi-year offtake | | Verizon / Google | Analyst-cited incremental projects | Soft positive for FTTH/DCI; management did not name-confirm | | Memory producers | Indirect (pricing power) | Better for memory makers; worse for smartphone unit volumes | | NA Class-8 OEMs | Diesel substrates | Sequential order recovery independent industrial data point | | U.S. solar domestic | Only U.S. poly + wafer maker | Better for domestic capacity; Q2 loss was maintenance, not demand |

Headline third-party stack: Apple / Meta / NVIDIA / Amazon LTAs are the core bullish read-throughs. Competitive framing is capacity scarcity and innovation-value pricing — no named optical peers.

Corning printed a clean double beat with EPS surprise re-widening, Enterprise Gen-AI optical re-kicked to +65%, and operating leverage still expanding (OM +190 bps, EPS +30%). The Q3 guide extends mid-teens sales / high-20s EPS growth with street EPS parked at the floor of the range — a setup that matches the company’s 10-quarter beat record. The honest caveats are narrative (acceleration language vs a ~100 bps slower YoY sales guide), CapEx intensity rising with the upgraded plan, and Photonics/Scale-Up remaining optionality outside the current P&L. Trajectory over absolutes favors the long book on optical AI infrastructure; prove it again on 2026-10-27.