Dr. Ayanna Howard: Why Ai needs a moral compass

Welcome back. Let's talk AI here on Fox Local. I'm Jasmine Cornell. Thanks for joining our conversation. AI, as you all know, is rapidly changing our world, but who's making sure it's working for everyone? A renowned roboticist, AI pioneer, and Spelman College president, Dr. Ayanna Howard says technology has no moral compass of its own and it's up For us, or up to us, that is, to provide one. So she joins us live now to talk about her new book and why the future of AI depends on all of us. Dr. Howard, thanks so much for joining us. Thank you for having me. Yes, so let's get into AI first. Will AI replace our jobs, or do we need to, or what should we do to make sure that this does not happen? Yes, so I truly believe that AI will transform and change our jobs guaranteed. It will not replace all of them. And so the things that we have to prepare for is to ensure that we can co-create and co-work with artificial intelligence, whatever the discipline, lawyer, filmmaking, science. Diversity among AI developers impact the technology that we're creating today. So when you think about this element of diversity, the fact is, is that humanity is very diverse, irrespective of age, gender, nationality, race, social economics. And so if we're designing artificial intelligence to improve the human quality, human life, it means all humans. And so this element of diversity in the coders and the community that's engaged and even in the outcomes of the artificial intelligence, we need to ensure that we have diverse lived perspectives that are working with artificial intelligence in order for it to work correctly. And talk more about why it's important to address the bias in AI now before it becomes built into our everyday lives. So one of the things about people is that we overtrust artificial intelligence. And so when AI is doing something, giving us a diagnosis with respect to health care, saying where you should go to college, the fact is, is we rely on that output. We basically overtrust it. And so if there is bias, if it has some. Unintended consequences sometimes in these elements, we're gonna overtrust it. So if a doctor is using AI, and the AI helps her with a certain diagnosis, gonna say, okay, but if the AI has not been trained on black women such as myself, so what is that gonna do with respect to my medical benefits and my outcomes, it'll be negative. Right now is the time that we have in order to advocate for the changes that are needed, to advocate for the different types of data, as well as the programmers, developers, our ethical community advocates in order to change it so that when it is out there, when it's pervasive, we will have some control over it. And, you know, you said that we as a society are putting too much trust into AI. How can we encourage people to set some boundaries when it comes to using artificial intelligence? So one of the things is that this boundary condition is going to be a little bit hard because we use AI in pretty much every application we have, even if we're not aware of it. But I always say, you know, an AI away day is a good day. It's like the apple a day. We have to be intentional in order to disconnect. We should say one day, you know what, I'm not gonna use those auto-suggested corrections in my email for this day and take a break. One, it allows ourselves, us to recalibrate and say, you know what, I do have my own critical thinking. I can think for myself. I don't have to rely on AI. It allows us to basically step back and take a pause and a breath. So important to disconnect at some point to not over rely and over trust AI. And let's talk about the growing use of robots here and how this can impact us negatively later in life. So the difference between what I would say physical AI, embodied AI, which is the physical robots and virtual AI, such as the chatbots, is that physical AI can change things in the world. It can cause real harm and real damage. Whereas a virtual chatbot, it can cause damage, but it's much more in terms of the emotional, the mental. So imagine that a physical robot is unable to identify my black skin versus someone else's skin color, and it's on a self-driving machine. Guess what, harm may actually come on my life, or not able to identify children or older adults, because just so you know, the training data doesn't have a lot of kids in the training corpus. So what does that mean if I have a healthcare robotic system that has no idea and is now actually having to do something in the hospital environment or in the clinic? These are damages that you can't replace. You can't say, oops, uh-oh. There goes someone's, that's harm, unlike we could do in the virtual AI. So robots, physical robots, has another level of concern. And I've been doing robotics since 1994. So I actually know what could happen if we don't really fix and really address these biases now. And Dr. Howard, our time is coming to an end here, but I do wanna ask you one more question. Let's get into your book and what it explores. So the rebooting the machines explores and it really provides a blueprint for anyone and this is whether you are a coder, you are a physician, you are a lawyer, or you're just out there wanting to explore AI. It is a blueprint a step by step of what every person can do in order to take control and empower yourself to actually have an impact in this world of AI. Dr. Howard, thanks so much for shedding light on this and being a part of our conversation today. We appreciate your time. Thank you. Thank you. And all book proceeds go to Spelman Scholarships, by the way. All righty. That sounds good. Thanks so much, Dr. Howard. Thank you.