Half of Meta's $1.9 Million Glasses Grant Fund Went to Five App Developers

Meta named the recipients of its AI Glasses Impact Grants in late July: 30 organizations across 18 states, awarded what the announcement rounds off as "nearly $2 million." The exact figure is $1,875,000. Sort it by grant type and the program splits almost perfectly in half.

The math

Accelerator grants went to organizations already using Ray-Ban Meta glasses in their day to day work. Meta funded 15 of those at $25,000 and 10 at $50,000. That is 25 recipients sharing $875,000, an average of about $35,000 apiece.

Catalyst grants went to app developers. There are five, at $200,000 each. Five recipients, $1 million, the larger half of the fund.

Close to 500 organizations applied for the 30 slots. The program opened in January and closed applications on March 9.

What the 25 are building

Eleven recipients work in accessibility and independent living. The list includes the American Council of the Blind, the Braille Institute of America, United Spinal Association, Deaf Equality, VisionCorps, the PACER Center, and Easterseals affiliates in Iowa and Greater Houston. The rest cluster around construction safety, field service, agriculture, and classroom education.

The specific projects are the part worth reading. OurLife Labs in California is building a dementia support assistant. The Mycelia Foundation is working on safety guidance for rural broadband installers in New Mexico. Agerpoint is doing crop health monitoring for farmers in North Carolina, Strategic Workforce Development is coaching construction trades hands-free in Missouri, and Proformance-AI is building a copilot for roadside mechanics.

Meta's own framing is the strongest case anyone has made for this hardware. When a job "requires both hands, full attention, and real-time information," the company wrote, glasses are easier and safer to use than a phone. Every project on that list is a version of that sentence.

What a Catalyst grant actually requires

The five big grants come with a technical condition attached. Catalyst recipients build on the Meta Wearables Device Access Toolkit, the SDK that lets an outside app talk to the hardware on a pair of Ray-Ban Metas. There is an iOS version in Swift and an Android version in Kotlin, both shipping with prebuilt libraries and sample apps.

What you can reach through it is narrow. You get the camera that captures the wearer's viewpoint, photo capture, and the microphone array and open-ear speakers. You do not get Meta AI. Voice commands are not exposed to third parties. And you cannot ship yet: publishing to the public is closed during the Developer Preview, with general availability promised sometime this year.

The eligibility language says something too. Catalyst applicants had to demonstrate "an existing mobile app with an established consumer audience." These grants were never pointed at a small organization with a good idea. They were pointed at companies that already have users, to bring those users onto the glasses.

The display arrived after the deadline

On March 9, when applications closed, the toolkit exposed cameras, microphones, and speakers on Ray-Ban Meta Gen 1 and Gen 2 and the Oakley Meta HSTN. Meta's grant FAQ still carries that generation of the language, noting the toolkit does not support in-lens display technology or the EMG wristband.

Then May happened. Meta opened Ray-Ban Display to third-party developers with two routes, a native mobile SDK and standalone web apps built in HTML, CSS, and JavaScript, both able to take input from the Neural Band. The current developer FAQ lists Ray-Ban Display as a supported device and documents Neural Band swipes and pinches inside Web Apps.

So the five funded developers applied against one platform and will build on a wider one. The flip side is that nothing in this cohort was proposed with the display in mind, because for outside developers it did not exist when the applications went in. The product that most needs third-party software got none of this money aimed at it.

Two things at once

Both readings hold up. The accessibility work is real, and $25,000 goes further at an organization the size of VisionCorps than the number suggests. It is also true that the single largest line item in a philanthropic program is a developer acquisition budget, paid to companies with existing consumer apps, to seed a store that has not opened.

Google ran its own version of this in May, handing out free Project Aura dev kits through the Android XR Catalyst program. Nobody called that philanthropy, and nobody needed to. Seeding a platform early is a normal thing for a platform owner to do. It is just worth reading the ledger alongside the headline.