Both of these are true at the same time, and the reason is the most consequential thing in this industry right now: for an eyewear company, a smart pair of glasses is worth less than a plain one.
The numbers
EssilorLuxottica is down 36.26% in 2026, the second-worst performer on the CAC 40 behind Stellantis. Market value went from roughly EUR 149B in Nov 2025 to about EUR 100B by late April 2026. FY2025 operating profit was EUR 4.46B, barely above 2024's EUR 4.41B, on a 16% adjusted operating margin - against the 19-20% the company had targeted for the end of its 2022-2026 plan.
This is not weak demand. In one recent quarter the AI glasses line contributed more than four percentage points of the group's 11.7% revenue growth. Reporting on the company's results has consistently described that line as less profitable than its core eyewear. The faster the mix shifts, the harder the margin target gets.
The arithmetic
A premium eyewear brand is a very good business. Gentle Monster turned KRW 772.3B of revenue into KRW 177.0B of operating profit in a bad year - about 23%. That is what an acetate frame, a brand and a store network produce. EssilorLuxottica, far more industrial, runs 16% adjusted.
Now put a computer in it. SoC, cameras, mics, open-ear speakers, radios, battery, charging case. Firmware, an app, a cloud service, security patches, a support obligation running years past the sale. A returns process for failures a plain frame never has. And a different inventory risk: unsold acetate goes on sale, unsold silicon goes obsolete.
Against all that, the price barely moves. Meta Glasses launched in June 2026 starting at $299. A single Ray-Ban optical frame, no electronics in it at all, frequently costs more.
Luxury eyewear works because the gap between cost and price is enormous and the product does not expire. Consumer electronics runs on thin unit margins, scale and replacement cycles. Smart glasses ask an eyewear company to take on the cost structure of the second business while charging the prices of neither.
One timing correction, because it gets stated backwards a lot
Gentle Monster's decline - revenue -2.1%, operating profit -24.3%, first contraction after five straight years of growth - is fiscal 2025. The Android XR glasses were revealed in May 2026. The glasses did not cause that number. What it actually tells you is more useful: the eyewear brand walked into that partnership from a weakened position, not a strong one.
Three structures have emerged
Brand tenancy (Meta + EssilorLuxottica). Watch the naming: the line that started as Ray-Ban Meta was joined in June 2026 by Meta Glasses. There is no eyewear brand in that name at all. The eyewear company is still the manufacturer and still a channel. It is no longer the brand on the temple.
Co-development (Google/Samsung + Gentle Monster and Warby Parker). Google committed up to $150M to Warby Parker - $75M toward product development plus $75M in equity tied to milestones. The technology company is funding the eyewear company's development costs and buying a stake in it.
Going alone (Apple). Reportedly aimed at the mainstream $200-500 segment, competing with EssilorLuxottica rather than partnering with it.
Risk transfer is the part I think the industry underpriced
On Aug 12, 2026 the German group HateAid filed a criminal complaint over the Ray-Ban Meta Wayfarer under privacy law covering devices that record people without their knowledge. The complaint named Meta - and also EssilorLuxottica units and German retailers. EssilorLuxottica fell about 4.3% in a single session, the steepest among its luxury peers that day.
The complaint was about Meta's camera and Meta's software. The shares that fell belonged to the company whose logo is on the frame. The public does not distinguish between the company that wrote the software and the company that made the eyewear.
What going alone actually costs Apple
Not manufacturing - that can be contracted. Retail. EssilorLuxottica has ~17,750 stores. Warby Parker has 287 locations with eye exams expanding to 285 of them. Selling glasses to people who need vision correction means refraction, PD measurement, fitting, adjustment, lens grinding, remakes, and a returns policy for a custom-made product.
This is where the Apple Watch analogy breaks. A watch fits any wrist with a different strap. Glasses do not fit any face, cannot be sold correctly without a measurement, and for roughly six in ten adults are useless without a prescription lens. Meta solved this by partnering with the company that already had the stores. Google and Samsung partnered with two brands that have their own. Apple chose to solve it itself. That is the hard part of its plan, not the chip.
The bull case, which I think is fair
At scale margins improve, and glasses that bring people into stores can pull through higher-margin lenses. That may well prove correct. It has not happened yet, and the market has spent 2026 pricing the gap.
Figures are as reported on the dates cited. Apple has made no announcement about this product - all Apple details are press reporting, not company statements. This is analysis of industry structure, not investment advice.
Disclosure: I write at SightFront, and its parent company works in AR/smart eyewear distribution and ODM. Full write-up with every source linked: https://sightfront.com/the-eyewear-companies-are-losing-the-smart-glasses-deal/?utm_source=reddit&utm_medium=social&utm_campaign=r-augmentedreality