Joy Buolamwini facts for kids

Quick facts for kids Joy Buolamwini
Joy Buolamwini speaking at Wikimania in 2018
Born	Joy Adowaa Buolamwini (1990-01-23) 23 January 1990 (age 36) Edmonton, Alberta, Canada
Education	Cordova High School
Alma mater	Georgia Institute of Technology (BS) Jesus College, Oxford (MS) Massachusetts Institute of Technology (MS, PhD)
Known for	Algorithmic Justice League Gender Shades study
Scientific career
Fields	Computer science Artificial intelligence Algorithmic bias
Institutions	MIT Media Lab

Joy Adowaa Buolamwini is a Canadian-American computer scientist, artist, and digital activist. She is known for discovering major fairness issues in artificial intelligence (AI) and facial recognition software.

Buolamwini founded the Algorithmic Justice League (AJL). This organization works to make technology fairer for everyone. She combines computer code, creative poetry, and scientific research to show the world how computers can make unfair decisions if they are not designed carefully.

Who Is Joy Buolamwini?

Joy Buolamwini calls herself a "poet of code." She is a scientist who uses both engineering skills and creative art to solve real-world problems.

Her research revealed that many commercial face-checking programs did not work well for people with darker skin tones, especially women. By speaking up, she convinced some of the largest technology companies in the world to change how they build and test artificial intelligence.

Today, Buolamwini is recognized worldwide as a pioneer in AI ethics and computer fairness.

Early Life and Growing Up

Joy Buolamwini was born on January 23, 1990, in Edmonton, Alberta, Canada. Her parents immigrated from Ghana, a country in West Africa.

When she was a young child, her family moved to the United States. She spent her childhood in Mississippi and later moved to Tennessee. She attended Cordova High School near Memphis, Tennessee.

Growing up in a home full of learning shaped her interests early on. Her father worked as a university academic, while her mother was an artist. This blend of science and creativity inspired Buolamwini to explore both computers and the arts.

Learning to Code at Age Nine

Buolamwini discovered her love for technology at nine years old. She watched a television story about Kismet, a famous robot built at the Massachusetts Institute of Technology (MIT). Kismet could move its eyes and show expressive emotions.

Seeing Kismet inspired her to create her own technology. She began teaching herself computer programming languages at home, including:

• XHTML (used for building webpage layouts)
• JavaScript (used for making websites interactive)
• PHP (used for running website servers)

She enjoyed building websites, solving technical puzzles, and testing new software ideas on her computer.

High School Years and Sports

In high school, Buolamwini was both a dedicated student and an active athlete. She competed in the pole vault track event and played on the school basketball team.

She often worked on challenging advanced physics homework during short breaks between basketball practices. Her hard work in academics and sports taught her how to manage time, solve tough challenges, and focus on long-term goals.

College and University Studies

Buolamwini attended the Georgia Institute of Technology (Georgia Tech) in Atlanta, Georgia. She studied computer science and focused on health informatics, which uses computer tools to improve medical healthcare.

While an undergraduate student, she became the youngest finalist in Georgia Tech's InVenture Prize contest in 2009. She graduated in 2012 as a top student under the Stamps President's Scholars program.

Studying Across the World

Buolamwini won several major scholarships to continue her higher education around the world:

• Fulbright Fellowship (2013): She traveled to Zambia in southern Africa. There, she worked with local computer scientists to teach Zambian youth how to build mobile applications.
• Rhodes Scholarship (2013–2015): She studied at Jesus College, Oxford at the University of Oxford in the United Kingdom. She earned a Master of Science degree focusing on learning and digital technology.
• MIT Media Lab (2015–2022): She joined the famous MIT research laboratory in Cambridge, Massachusetts. She completed her master's degree in 2017 and earned her Doctor of Philosophy (PhD) degree in 2022.

How Joy Buolamwini Discovered AI Bias

While working as a graduate researcher at the MIT Media Lab, Buolamwini worked on creative art inventions. One of her projects was a digital mirror called the "Aspire Mirror."

The Aspire Mirror used a web camera and facial-analysis code. It was designed to project images of inspiring historical figures onto the user's reflection in real time.

The White Mask Experiment

When Buolamwini sat down in front of the camera, the software failed to detect her face. Because she has dark skin, the program simply could not tell a human face was there.

To test the system, Buolamwini placed a plain white plastic mask over her face. As soon as she put on the mask, the computer immediately detected a face.

She realized the computer vision model had not been trained on enough diverse images. Because the software was tested mainly on lighter skin tones, it failed when looking at someone with dark skin.

What Is the "Coded Gaze"?

Buolamwini created the phrase coded gaze to explain this problem.

The coded gaze describes how the personal experiences, assumptions, and choices of software creators get programmed directly into computer algorithms. If computer engineers only test software on people who look like themselves, the machine will fail when used by a wider public.

The Landmark "Gender Shades" Study

In 2018, Joy Buolamwini teamed up with computer scientist Timnit Gebru to conduct a scientific investigation titled Gender Shades.

How the Scientific Test Worked

Buolamwini and Gebru wanted to see if facial-analysis programs sold by top tech companies worked equally for everyone. They tested commercial systems created by major companies:

• IBM
• Microsoft
• Face++ (a large computer vision firm)

The researchers built a new scientific dataset called the Pilot Parliaments Benchmark. It contained 1,270 diverse faces of government leaders from three African countries and three European countries. This ensured a balance of skin tones and genders.

Shocking Results

The study measured how accurately each software system could classify a person's gender from a photograph. The findings surprised the tech industry:

• For lighter-skinned men, the error rate was extremely low, often less than 1%.
• For darker-skinned men, the error rate increased slightly.
• For lighter-skinned women, the systems made a few more errors.
• For darker-skinned women, the error rate reached up to 34.7%, and sometimes misclassified faces nearly half the time.

Why These Errors Occur

Computers learn to recognize patterns using machine learning. Programmers feed millions of photos into an algorithm so it learns what a human face looks like.

Buolamwini discovered that older AI image sets were over 75% male and more than 80% lighter-skinned individuals. Because the systems saw very few darker female faces during training, they never learned how to identify them accurately.

Changing the Technology Industry

The Gender Shades study became one of the most influential computer science papers of the 2010s. It forced large tech companies to re-examine their products.

Corporate Policy Changes

• IBM: Updated its facial recognition algorithms within months, reducing error rates significantly. In June 2020, IBM stopped selling facial recognition software entirely.
• Microsoft: Re-trained its artificial intelligence models with diverse datasets to fix accuracy gaps.
• Amazon: In 2020, Amazon paused the sale of its facial recognition tools to law enforcement agencies.

Buolamwini proved that independent research could hold powerful corporations accountable.

The Algorithmic Justice League (AJL)

In 2016, Buolamwini founded the Algorithmic Justice League (AJL). The organization brings together researchers, artists, and community organizers.

Core Goals of the AJL

The Algorithmic Justice League focuses on three major goals:

1. Highlight AI Bias: Exposing algorithmic discrimination through scientific testing and artistic storytelling.
2. Promote Accountability: Encouraging companies and governments to check their software for hidden flaws before releasing it.
3. Empower the Public: Teaching regular citizens how automated tools affect their daily lives.

Creative Campaigns and Educational Projects

The AJL uses creative methods to teach people about artificial intelligence:

• The Safe Face Pledge: A pledge asking technology companies not to build surveillance tools that infringe on human rights.
• Voicing Erasure: A digital media project highlighting how speech-to-text systems frequently misunderstand certain regional accents and voices.
• CRASH Project: A community system allowing everyday users to report errors and harms caused by automated programs.
• STEM Education: Partnering with youth groups like Black Girls Code to teach young students computer science skills.

Media, Books, and Public Speaking

Joy Buolamwini shares her research through movies, public speeches, and books to reach audiences outside the scientific community.

Documentary: Coded Bias

Buolamwini's research is the central focus of the 2020 documentary film Coded Bias, directed by Shalini Kantayya.

The film follows Buolamwini from her laboratory at MIT to public hearings in government halls. It shows how automated software is used in housing, employment, and public spaces. The documentary premiered at the Sundance Film Festival and became available worldwide on Netflix, earning an Emmy nomination.

Book: Unmasking AI (2023)

In 2023, Buolamwini published her book, Unmasking AI: My Mission to Protect What Is Human in a World of Machines.

In the book, she shares her personal journey as a young coder and explains complex technological concepts in clear language. She calls for:

• Inclusive training data for all AI models.
• Independent testing of software before it enters public use.
• Clear laws to protect people from unfair automated decisions.

Public Policy and Government Work

Buolamwini has spoken before government leaders to explain how artificial intelligence works:

• U.S. Congressional Testimony (2019): She testified before the United States House Committee on Oversight and Reform regarding the risks of unregulated facial recognition.
• White House Consultations: She advised government officials on creating safety rules for new AI systems, contributing to discussions around federal AI safety guidelines in 2023.

Awards and Honors

Joy Buolamwini has received many awards for her scientific achievements and public leadership:

• Search for Hidden Figures Grand Prize (2017): A national award honoring emerging women leaders in science, technology, engineering, and math.
• BBC 100 Women (2018): Named one of the one hundred most inspiring and influential women in the world.
• Fortune's World's 50 Greatest Leaders (2019): Listed among top global leaders and described as the conscience of the AI revolution.
• Time 100 Next (2019) and Time 100 AI (2023): Recognized by Time magazine for shaping the future of artificial intelligence.
• Honorary Doctor of Science (2024): Awarded an honorary doctorate degree by Dartmouth College.
• NAACP Archewell Digital Civil Rights Award (2024): Honored for creating fairer digital spaces for diverse communities.

Key Ideas Explained

Understanding Buolamwini's research helps explain how modern technology functions:

What Is an Algorithm?

An algorithm is a step-by-step list of instructions given to a computer to solve a problem or complete a task. It is similar to a cooking recipe written in computer code.

What Is Algorithmic Bias?

Algorithmic bias happens when a computer system creates systematically unfair results, such as favoring one group of people over another. This usually occurs because the data used to train the system was incomplete or unbalanced.

What Is Training Data?

Training data is the large collection of examples, such as photos, text passages, or audio recordings, that engineers feed into a machine learning program so it can learn how to perform a task.

Images for kids

• 

  Joy Buolamwini speaking to an audience at Wikimania 2018 in Cape Town, South Africa

• 

  Logo of the Algorithmic Justice League, founded by Joy Buolamwini

Related Pages

In Spanish: Joy Buolamwini para niños

• Artificial intelligence
• Algorithmic bias
• Facial recognition system
• Computer vision
• Ethics of artificial intelligence
• Timnit Gebru