Short Wave Are You Overestimating The Algorithm?
Sep 3, 2024
Joy Buolamwini, a computer scientist and author of "Unmasking AI," discusses the often misunderstood nature of algorithms. She and host Kyle Chayka tackle the concept of AI 'hallucinations,' explaining why they're more than just quirky errors. They delve into the ethical implications of facial recognition technology and highlight the biases that can arise from flawed algorithms. The conversation also touches on how algorithms influence personal relationships and the urgent need for advocacy in shaping a more equitable technological landscape.
AI Snips
Chapters
Books
Transcript
Episode notes
Algorithms: Then and Now
- Algorithms are fundamental equations used to achieve desired outcomes.
- They've existed since ancient times, initially for calculations like crop division or ditch volume.
Algorithms as Instructions
- Algorithms are sequences of steps to achieve a specific result, like moving a character in a video game.
- This is a basic form of communicating with computers to program them for problem-solving or expression.
Facial Recognition's Flaws
- Joy Buolamwini's work reveals the serious side of AI algorithms, particularly in facial recognition technology.
- This technology has flaws, including racial bias, leading to misidentification and wrongful arrests.
