Lightwave Logic, Inc. (LWLG) | The Buildout — AI Infrastructure
The Verdict
Lightwave Logic makes Perkinamine, an electro-optic polymer material engineered to sit on top of a silicon photonics chip and make its optical modulator faster, smaller, and lower-power. The modulator sits within the optical link, and the network layer is where AI clusters hit a limit — as management puts it, the network has to keep up with the compute. The company does not sell a finished component. It sells or licenses the material and device designs into foundry-built silicon photonics, aimed at the three interconnect markets it names as scale-up, scale-out, and scale-across.
| Market Cap | — |
| Revenue (TTM) | $0M |
| EBITDA Margin (TTM) | -11000.0% |
| Net Cash | $93M |
| P/E (TTM) | — |
| EV/EBITDA (TTM) | — |
What We Like
- Five Stage 3 customer programs as of the Q2 2026 call, up from four; the addition is an unnamed Fortune Global 500 program initially focused on scale-across transceivers using coherent modulation.
- One material supply and licensing agreement is described as already in place, and a second is under active negotiation with the customer management calls the most advanced on the path to productization.
- Three dedicated foundry runs underway inside a four-partner ecosystem: two wafer sets were expected in August 2026, the third foundry in Q4 2026, and a fourth tape-out was targeted for later in 2026.
- $95.9M in cash, cash equivalents, and marketable securities at June 30, 2026, against a pre-commercial revenue base.
- Latest-generation materials passed Telcordia-related stress testing including 85°C and 85% relative humidity, reported on the Q1 2026 call and not refreshed in Q2.
What We’re Watching
- No customer program has ever reached Stage 4, the manufacturing/qualification stage, and every revenue-bearing commitment sits at or beyond H2 2027.
- Foundry capacity remains the binding external gate: management says those pressures 'have not disappeared,' and that schedules, design changes, packaging, and system testing can all affect the pace.
- Cash burn is running ahead of revenue — about $9.9M used in operating activities in H1 2026 versus $7.3M a year earlier; the Roth sales agreement was raised to $51,404,500 with only $3,385 remaining at the 10-Q filing date.
- Concentration: two unnamed customers were 55% and 45% of FY2025 net sales, and 100% of Q1 2026 net sales were billed to Switzerland.
The thesis rests on converting a staged pipeline into qualification and signed supply deals, and the Q2 2026 evidence shows steady engineering progress rather than a commercial inflection. Pipeline counts widened, foundry wafers got dates, and a first signed commercial framework was described as in place. Against that, revenue is still about $33,000 for the quarter, R&D and G&A rose, no program has reached qualification, and the back-end-of-line step that gates volume still has no named partner, site, capacity, or capital figure. The open question is whether the Q4 2026 wafer and characterization results meet customer requirements — the event management itself calls the point that 'should give us a much clearer technical picture.'
Earnings
Q2 2026 revenue was approximately $33,000, up from roughly $26,000 a year earlier and still de minimis; gross margin on that base is not a meaningful signal. The net loss widened to $6.6M from about $5.0M a year earlier. R&D rose to about $3.9M from $2.6M, and G&A to about $3.4M from about $2.3M. The company ended June 30, 2026 with $95.9M in cash, cash equivalents, and marketable securities, and headcount rose 25% sequentially in manufacturing scale-up, test engineering, and commercial development.
| Metric | Q2 FY2026 | Q1 FY2026 | Q2 FY2025 | YoY |
|---|---|---|---|---|
| Revenue | $0M | $0M | $0M | — |
| Gross margin | 100.0% | -1790.1% | 86.5% | +1350bps |
| EBITDA | −$7M | −$6M | −$5M | +31.4% |
| EPS | $-0.04 | $-0.04 | $-0.05 | −5.5% |
| Stage 3 customer programs | 5 | 4 | n/a | — |
| Deferred revenue (JDA, milestone-gated) | ~$100,000 | n/a | n/a | — |
One such material supply and licensing agreement is already in place.— Yves LeMaitre, CEO, 2026-08-11
Management tone: Management's commentary shifted from pipeline counts toward specific dates and a signed agreement. The Q1 2026 call leaned on Stage 3 counts and a TAM revision; the Q2 2026 call substituted an August wafer window, a Q4 third-foundry delivery, a fourth tape-out later in 2026, and one material supply and licensing agreement described as in place. Management repeatedly tied qualification timing to factors it does not control and declined to claim Perkinamine is the only viable material for co-packaged optics. Both calls used investor-submitted, IR-read questions rather than live analyst Q&A.
Management Guidance
No formal financial guidance has been provided on any call in the record. The Q1 2026 10-Q states that revenue, if any, recognized during 2026 would primarily relate to material supply, NRE arrangements, or prototype and development activities, and that the company does 'not currently expect significant revenue from volume commercial production of customer products until 2027 at the earliest.' On the Q2 2026 call management narrowed the goal for the most advanced program to beginning volume production in the second half of 2027. The same filing says cash finances operations at least through December 2027; on the Q2 call, a direct runway question was answered with a balance-sheet statement rather than a timeline.
Trajectory
Revenue has been effectively zero for years and remains so: $29,167 in Q1 2026 and about $33,000 in Q2 2026, against roughly $26,000 a year earlier, on TTM revenue of $0.2M. What is moving is cost. R&D rose to about $3.9M from $2.6M, and G&A to about $3.4M from about $2.3M; TTM EBITDA is -$22.0M, net income -$22.8M, and free cash flow -$18.2M. H1 2026 operating cash use was about $9.9M versus $7.3M a year earlier, with H1 capex of about $1.5M. The operating story is widening rather than inflecting: Stage 3 programs went from 4 to 5, and foundry engagement now runs across four partners.
The Model
The model projects FY+1 revenue of $0.3M with EBITDA of -$27M (-9000%), and FY+2 revenue of $2.5M with EBITDA of -$31M (-1200%). The near term is anchored by the company's own filing language: revenue in 2026, if any, would come from material supply, NRE, and prototype and development activity rather than volume production. The FY+2 step up is tied to the goal of beginning volume production for the most advanced customer program in the second half of 2027, which depends on wafer results, qualification, and a second licensing agreement. Dispersion is wide — across the five runs, FY+2 revenue ranges from $1M to $8M.
| Metric | FY2025 | Next FY (E) | Following FY (E) |
|---|---|---|---|
| Revenue | $0M | $0M | $2M |
| YoY Growth | — | +50.0% | +733.3% |
| EBITDA | −$19M | −$27M | −$31M |
| EBITDA Margin | -9300.0% | -9000.0% | -1200.0% |
Projections are the median of 5 independent model runs.
No formal financial guidance has been provided on any call in the record. The Q1 2026 10-Q states that revenue, if any, recognized during 2026 would primarily relate to material supply, NRE arrangements, or prototype and development activities, and that the company does 'not currently expect significant revenue from volume commercial production of customer products until 2027 at the earliest.' On the Q2 2026 call management narrowed the goal for the most advanced program to beginning volume production in the second half of 2027. The same filing says cash finances operations at least through December 2027; on the Q2 call, a direct runway question was answered with a balance-sheet statement rather than a timeline.
What Could Go Right — and Wrong
- Wafer results in Q4 2026 meet customer performance requirements in customer-relevant foundry flows, moving at least one program toward Stage 4.
- The second material supply and licensing agreement is signed, giving a second commercial framework alongside the one already in place.
- A Stage 3 program reaches Stage 4 and qualification — something that has not happened for any program.
- The back-end-of-line volume path becomes redundant, with a named partner or a second line removing the single point of failure.
- Foundry capacity and cycle times ease, validating the Q1 2026 expectation that constraints would gradually disappear over the next 12 months.
- Wafer or characterization results miss customer requirements or require material and design changes, pushing qualification and revenue further out.
- Foundry capacity, process, yield, packaging, or test constraints persist — 'those pressures have not disappeared' — extending cycle times and pushing the H2 2027 target out.
- A competing platform wins a key Stage 3 design; the 10-K names silicon-based modulators, thin-film lithium niobate, and III-V/indium phosphide.
- The second licensing agreement stalls, leaving one signed framework with no disclosed counterparty and no disclosed economics.
- Cash burn outpaces the runway and forces further equity issuance; the Roth sales agreement had $3,385 remaining at the 10-Q filing date.
Looking Ahead
The next twelve months run through the foundry. Two wafer sets were expected in August 2026, the third foundry was expected to deliver in Q4 2026, and a fourth foundry was preparing a run with a tape-out targeted for later in 2026, followed by back-end-of-line processing and testing. Management calls the fourth quarter the point that should give a much clearer technical picture and set up the next decision with those customers. The dated pipeline target is 1 to 2 additional Tier 1 customers reaching Stage 3 before the end of 2026, and one has landed so far. The commercial trigger is the second material supply and licensing agreement, with volume production for the most advanced program targeted for the second half of 2027.
- August 2026Foundry wafer deliveries — Two wafer sets expected from two dedicated foundry runs.
- Q4 2026Third foundry wafers — Third dedicated foundry run expected to deliver its wafers.
- Q4 2026Technical readout — Wafer processing, testing and characterization; management sees a clearer picture.
- Later in 2026Fourth foundry tape-out — Fourth foundry dedicated run being prepared; tape-out targeted this year.
- End of 2026Stage 3 additions target — 1 to 2 additional Tier 1 customers targeted; one has landed.
- H2 2027Volume production goal — Target for the most advanced customer program, per management.
Financials
Annual Summary
| Metric | FY2024 | FY2025 | TTM | YoY |
|---|---|---|---|---|
| Revenue | $0M | $0M | $0M | — |
| Gross Margin | -263.0% | -453.4% | -228.4% | 19,038bps |
| EBITDA | −$21M | −$19M | −$22M | +12.7% |
| EBITDA Margin | — | -9300.0% | -11000.0% | — |
| Net Income | −$22M | −$20M | −$23M | +9.8% |
| Free Cash Flow | −$18M | −$15M | −$18M | — |
| Net Cash | — | — | — | — |
Key Ratios (Trailing)
- P/E TTM—
- EV/EBITDA TTM—
- EV/Revenue TTM—
- Price/FCF TTM—
- Gross Margin (TTM)-228.4%
- EBITDA Margin (TTM)-11000.0%
- Net Margin (TTM)-11400.0%
- ROIC-177.8%
- SBC / Revenue550.0%
The Company
Lightwave Logic is a specialty materials and intellectual property company. Its product is the Perkinamine family of electro-optic polymer materials, engineered for integration into silicon photonics and other photonic integrated circuit platforms. The material is designed to make the optical modulator — the component that turns electrical signals into light — faster, smaller, and lower-power. Management frames the technology as complementary rather than competitive with silicon photonics: 'We are not asking the market to build completely new manufacturing systems around us. We are working to make the silicon photonics platform better.' The company reports a single reportable segment and a single product family.
The company operates from a single facility at 369 Inverness Parkway, Suite 350, Englewood, Colorado — 23,104 square feet, including a 1,000 sq ft class 1,000 cleanroom, a 500 sq ft class 10,000 cleanroom, chemistry laboratories, and analytic laboratories. All long-lived assets are located in the United States, totaling $9,411,107 at March 31, 2026. Manufacturing runs in three steps: Perkinamine production in Denver, which is being expanded; front-end silicon photonics PIC production at partner foundries; and back-end-of-line polymer deposition and encapsulation, done in-house in Colorado today. The company plans to monetize through material sales, IP licensing, process design kit enablement, and royalty or other fee-based arrangements tied to customer production.
Business Segments
Competitive Landscape
The competitive field is broad, and the company describes it that way. The FY2025 10-K names three competing categories: conventional silicon photonics modulators that rely on carrier depletion or carrier injection effects within silicon waveguides; thin-film lithium niobate, which leverages the electro-optic properties of crystalline materials to achieve high-speed modulation; and III-V compound semiconductors such as indium phosphide used in certain modulated laser and photonic device architectures. Management positions EO polymer as a layer that can address scale-up, scale-out, and scale-across 'unlike other technologies such as indium phosphide or TFLN,' but declined to claim Perkinamine is the only viable material for co-packaged optics: 'Competition is fierce, and the industry's brightest minds are assessing all kinds of materials and solutions.'
- Silicon-based modulators (carrier depletion / injection)Named in the 10-K as a competing category; described as relying on carrier depletion or carrier injection effects within silicon waveguides.
- Thin-film lithium niobate (TFLN)Named in the 10-K; described as leveraging the electro-optic properties of crystalline materials to achieve high-speed modulation.
- III-V compound semiconductors (indium phosphide)Named in the 10-K; described as used in certain modulated laser and photonic device architectures.
- HyperLightAppears in the supply-chain wiring set as a TFLN modulator competitor; not discussed in the filings.
- NLM PhotonicsAppears in the wiring set as an EO polymer modulator competitor; Tower Semiconductor's prior-quarter call named NLM Photonics alongside Lightwave Logic as an organic-polymer partner.
Supply Chain
Lightwave Logic sits upstream in the silicon photonics chain: its polymer is deposited onto modulator chips built at partner foundries, and its customers are the chip and module makers designing those modulators. None of the named foundries mentioned Lightwave Logic by name in their latest quarterly calls.
More on LWLG: Earnings recap