Artificial Intelligence Needs Diversity to Prevent Society’s Biases

แชร์
ฝัง
  • เผยแพร่เมื่อ 18 พ.ย. 2024

ความคิดเห็น • 3

  • @janedeuber1596
    @janedeuber1596 7 หลายเดือนก่อน

    Speak the truth, Nelly. As a woman in the tech world I so appreciate you voicing this. This insight needs to go viral!!!

  • @zenpoolbuddha1787
    @zenpoolbuddha1787 2 ปีที่แล้ว

    Lets!! GOOOOO!!!! #TeamZenPool Loves Nelly Yusapova's Content! Can't wait to connect and share your message with the Billiard Players of the World!!!

  • @simsimmons8884
    @simsimmons8884 2 ปีที่แล้ว

    Hello. Having worked in robotics since 2008, I believe I have some knowledge worth adding to this video. The first thing I would like to say is thank you for trying to encourage women to get into the tech fields and AI. It is clear that AI is going to be critical for many aspects of society over the next 20 to 30 years, and longer in many different forms.
    My problem with some of the things that you said was that men put their biases into the AIs that they create. I would suggest that that is a very simplex understanding of AI, the data sets, and the training that goes on to any AI or machine learning device. There's no question that data sets and the way AIs and machine learning systems are designed leaves a large space for unknowns that they're not trained on. Instead of suggesting this is a male versus female thing, I would suggest it's just the fact that we're early in understanding how to design and train AI systems.
    I had a discussion many years ago with a group of roboticists in 2010 about this very subject and the fact that we need to design wisdom into our robotics and AI systems. The problem is it's easier to state the problem than to fix the problem.
    I am writing my next book on AI architecture to achieve consciousness and I have quite a big section on ideas behind adding wisdom to our systems. If we don't think about this as we design our systems, it will just be one more problem that machine learning and AI systems will always have.
    As an engineer and not a neuroscientist, I look at the ways to build things such that they meet all the criteria of the design. I appreciate any discussion on bias, problems, and adding wisdom to our AI friends that will be with us in 10 or 20 years. I can't wait.