As companies look to mitigate racial bias in AI, one startup is looking to fix the problem by building a new model trained on a more diverse set of cultural and historical data.
Latimer, named after the 19th-century Black inventor Lewis Latimer, is training a new large language model (LLM) to more accurately represent Black history and culture. According to Latimer founder and CEO John Pasmore, the idea came to him earlier this year when he noticed no other companies were approaching racial bias within LLMs as a technical issue to solve.
“Not only was I seeing issues of bias, but there were inaccuracies in the history,” Pasmore said. “I have a 16-year-old, and the last thing you want is history to be baked incorrectly into what I consider the next round of [technology] just as big as search.”
Continue reading this article on digiday.com. Sign up for Digiday newsletters to get the latest on media, marketing and the future of TV.