Growing up, Angela F. first noticed gender bias in research materials when writing a third-grade report.

Vedanuj G. would like to share his grandmother’s poetry, but it’s written in a language not often understood outside her home state in northeast India.

Former math teacher and planetarium director Monica A. has seen firsthand the powerful impact immersive learning experiences can make in people’s lives.

Aigerim S. is passionate about giving people options to represent themselves in the metaverse exactly how they want to, from physical traits to clothing styles to assistive devices.

What do Angela, Vedanuj, Aigerim and Monica have in common? All four have brought their personal dedication to diversity, equity and inclusion (DEI) to Meta, where they’ve worked with colleagues and external partners to build technologies and programs—at Meta and beyond—that are shaping the future of technology in the reflection of and in service of the people who rely on it.

Angela F. recalls a third-grade school assignment: write about a historically significant person. The one requirement was that there be a book about that person in the local library. Although Angela was eager to write about Eleanor Roosevelt, she could find only a biography of Eleanor’s uncle, Teddy.

By the time Angela was in grad school, she found that a similar gender bias had made its way into the first place that school kids now typically look when given a research topic: Wikipedia.

The Wikipedia Foundation itself has found that as few as one in five English-language articles on Wikipedia are about women—a lack of representation that also includes non-binary people, members of the LGBTQ+ community and those outside the Western world.

“The world around me is a lot more diverse than what I see on Wikipedia,” says Angela, now a New-York-based research scientist for Meta AI.

Angela is open-sourcing an end-to-end AI model that can research and write first drafts of Wikipedia-style biographical entries for noteworthy individuals who are currently not on the site. Angela hopes this work, building off her PhD project, can eventually help Wikipedia writers and editors spotlight marginalized and intersectional groups.

In her current job, Angela appreciates the freedom Meta provides to turn her passion for diversity of representation into meaningful research. “I really want to work on things that speak to the problems that I’ve personally experienced,” she says.

Vedanuj G.’s grandmother is a poet. But her spoken and written language, Assamese, is seldom understood outside her home state of Assam, in northeast India.

“I envision a future where I can easily translate my grandmother’s poetry so that the world can appreciate it,’’ says Vedanuj.

Vedanuj, an engineer, is helping make that future possible. He’s part of a Meta team of AI researchers working toward better machine-translation tools. Their goal: Improve translation for the estimated 50 percent of people around the globe for whom the few dominant languages of the internet prevent understanding or otherwise connecting with most people of the online world.

“Language is everything,” points out Safiyyah S., a technical program manager. “It is the way we are able to express ourselves.”

That’s why Vedanju, Safiyyah and their Meta teammates have created a program to develop high-quality machine translation capabilities for most of the world’s languages. With their latest advancement, they’ve built an AI model that can now translate 200 different languages—including 56 African languages, which have historically been underrepresented in translation efforts—with state-of-the-art results, allowing more people to access online content.

With more than 7,000 different languages spoken on our planet, it will take much more work to provide everyone around the world with truly universal, inclusive translation tools. But by open sourcing their code and research tools the Meta team is enabling other researchers around the world to join the effort.

“The idea is to get rid of language barriers altogether," says Necip F.A.. “The metaverse is where all these things will come together.”

Aigerim S. says the metaverse is still in the early days of development. But she’s enthusiastic about the progress Meta has made to give people more ways to express and represent themselves with Meta Avatars.

“Our vision of the metaverse is to reach a billion people over the next decade,” Aigerim says. That means making people feel included and represented—whatever their country, ethnicity or language, or whatever their differing abilities and disabilities.

“People should be able to create avatars that speak to how they’re feeling and their authentic selves—whatever that means to them,” says Aigerim, who is originally from Kazakhstan.

So far, Meta Avatars give people more than one quintillion different combinations of options. Those options include things like skin tone, eye color, eyebrow shape, hair styles and more—along with devices like hearing aids or wheelchairs.

No one company will single-handedly create the metaverse. Aigerim’s team is collaborating with a broad range of entities—including the Organization of American States, Africa No Filter, Seoul National University and the United Spinal Association.

“We want the feedback from the community, ” she says, “so we can build in optionality and diversity from day one.”

The goal, in contrast to the way the internet was originally developed, is to design diversity and inclusion into the metaverse from the outset. This also means ensuring that the people building the metaverse come from a variety of backgrounds.

Throughout her career, Monica A. has been devoted to inclusive, immersive education—whether as a middle- and high-school math teacher in Boston or as director of a planetarium, where she developed an astronomy curriculum for K-12 students.

Now, as Head of Immersive Learning at Meta, she is leveraging her experience to lead the company’s $150 million investment to help develop the next generation of creators, fund immersive experiences and increase access to learning through technology.

Bringing experiences to people wherever they are through AR and VR can open new economic opportunities for all, and can make education more accessible to historically excluded learners, like those living in remote locations or people with limited mobility. Immersive learning can also open new economic opportunities to people who might not traditionally have had access to leading-edge skills training.

Meta partners with nonprofits, educational institutions, community groups, libraries and museums to demonstrate the possibilities for learning in the metaverse. One example is the Meta Spark curriculum in partnership with the Organization of American States to support the AR creator ecosystem in Latin America.

Another collaboration is Meta’s partnership with Victory XR, which is providing digital-twin campuses at a growing number of colleges and universities, including HBCUs like Morehouse College. These metaverse campuses, in fully spatial 3D, are designed to let students remotely access classes, social settings and extracurricular activities.

“Through virtual and immersive experiences in the metaverse,” Monica says, “we can increase access to education and help build a more connected and curious world.”