Jenny Lay-Flurrie explains why AI training data needs to include people with disabilities. Major tech companies partnered with the University of Illinois to improve speech recognition for disabled users, whose voices are often missing from datasets that make these systems work.
I think we all know that artificial intelligence is based on data. It's based on a lot of data, billions of data points that then give you the ability to have machine learning and intelligence coming out of it. It's got to be ethically done, it's got to be thoughtful, it's got to be inclusive. It's also got to include data that's come from people with disabilities in that mask, because if it doesn't, you don't have an end result that's going to actually empower the full diversity of human. So this week we just announced a great collaboration which we've been working on for some time with four other tech companies, small ones, Apple, Google, Amazon, Meta small ones to work with the University of Illinois to, with the right security, privacy, gather that data, working on impacted voice, so speech recognition, so that we can accelerate and advance quickly the technology that's available to people who need that. The fact that you've got five huge companies working together to accelerate a technology area to really lean into AI potential that get. Yeah, that gets me up in the morning.
O processo de reconhecer limites, mesmo quando a vida parece exigir sempre mais.
Depois de tantas histórias, aprendizados e pausas necessárias, o @interior_iza está de volta
