“In AI, Human Goodness Matters”: Julie Shin Choi with Intel AI (Video + Transcript)

Transcript of Elevate 2020 Session

Rachel Jones: Doing our afternoon keynote is Julie Shin Choi. Julie is the Vice President and General Manager of artificial intelligence products and research marketing at Intel Corporation. Prior to Intel, she led product marketing at HP, Mozilla, and Yahoo. So we are so excited to have Julie’s expertise this afternoon. She will be talking about how, in AI, human goodness matters. So welcome Julie.

Julie Shin Choi: Thanks so much. So let me just… I do have some slides. All right everyone. It is really good to be here with you today. Thank you so much for the intro. I am so glad to be here. It is Women’s History Month and in two days we’ll be celebrating International Women’s Day and what better way than to be together here at Girl Geek X Elevate Virtual Conference. Thank you so much to the Girl Geek X team and everyone behind the scenes for giving us this platform and this opportunity to connect. So let me share a little bit about myself. I am a VP at Intel, responsible for AI marketing, but really it’s been a long journey to get to this point. I absolutely love the job that I’m in and I thought it would be good to just share a little bit about that journey.

Julie Shin Choi: It’s been a 20 year career in tech, so far, mostly in Silicon Valley. I started my career in Boston and moved to the Valley in 2003, I think, so I’ve been here for about 17 years. When I thought about how my career has unfolded, I created this two by two, basically dividing the way I’ve been focusing my energy over the past 21 years. And as you can see, it’s an interesting graph that shows roughly 50% of the energy has been around life and 50% has been around career, and there’s different peaks and dips in how much I spend on each of these portions. But what’s been fascinating is that life and career converge in the same space time continuum, and it’s been an incredible journey. What I’ve learned about career and life is that it really is a series of choices that lead to opportunities and that these opportunities ultimately have led me to personal and career learning and growth.

Julie Shin Choi: So a very important choice that I made in 2016 was to join Intel, and joining Intel really did… It was a rocket ship moment for me. By that point, I joined Intel, I was a director at HPE prior to Intel. That was an amazing time as well. In between HPE and Intel, I met a startup called Nirvana and the CEO and the founding team of that startup and they got acquired into Intel and asked me to help do this AI thing at Intel. I did not know much about Intel other than it was the giant of computer processors and a hardware company and it was just too much of an opportunity to pass up. And so that choice was really profound because it led me to begin the AI journey with Intel. And we’ve come such a long way in the past nearly four years and we have evolved to really understand our place in the AI universe.

Julie Shin Choi: This is just a small slide. I want to thank Banu for the excellent talk she just gave. I had a chance to listen in and I’m just going to say a little bit because she did such a great explanation, but Intel AI, you can think of it as all the parts that you would need as a technologist, from hardware to enabling software to the memory, storage, and fabric. So many components go into building AI, and Intel is massively passionate and committed to building those components so that we can power this AI evolution and transformation across virtually every industry. But one of the reasons that I chose Intel was the opportunity and the scale that this technology platform would provide from a career perspective, but I did not anticipate that I would also fall in love with the people of Intel.

Julie Shin Choi: It is really this human goodness at Intel that keeps me here. At Intel we are really building technology to enrich the lives of every person on earth, that’s what Bob says, and I really believe that and I think it’s for the team that I remain, and it’s an incredible team. So let’s talk about some of the work that the team is doing. The title of this talk was “AI and the Importance of Human Goodness,” and one of the things that we’ve learned over the past three years is that AI is a powerful agent for helping people around the world, and this example comes to us from the Red Cross. We shared this example earlier this year at CES. Bob actually talked about it and there’s a video and I will tweet the video out after this talk, but basically this partnership is between Red Cross. Everyone knows Red Cross, it’s just an amazing relief organization dedicated to helping people in times of disaster.

Julie Shin Choi: And this partnership between Intel and Red Cross, as well as Mila, which is an AI think tank in Montreal, and other organizations, basically it was a data science partnership alliance, and the end result and objective was to map unmapped parts of Uganda and to identify, through deep learning, different bridges that relief agency Red Cross could take in times of disaster. At the end of the day, we were able to examine huge satellite images and come up with algorithms that could automatically identify the bridges, over 70 bridges in Uganda. So this is our first example of why human goodness matters when we think about AI application development. The second example is a little bit more current and relevant. I’m sure everyone has heard about and is taking precaution against the coronavirus epidemic that’s going on globally. And basically, what’s important is to use AI right now. Globally, we’re using big data, we’re analyzing different databases of where people have gone and the different symptoms that they may present.

Julie Shin Choi: But one novel use case that we identified in Singapore is of a company that’s using IoT technology to help scan people and identify thermal readings, so basically fevers, without human contact. And this is proving to be about three to four times more efficient, so we can scan 7 to 10 people with this AI device, as compared to using human healthcare practitioners. So in this way AI is really helping manage a lot of the issues related to coronavirus in Singapore. And we see other innovations like this cropping up all around the world. Another example of the intersection of AI and human goodness can be found in a collaboration between a company named Hoobox and Intel. Hoobox is a really fascinating company based in Latin America with North American operations as well, and they are dedicated to robotics for helping people with mobility issues and other novel uses of computer vision to aid humans, and this use case is a fascinating one where we collaborated with Hoobox. Intel provided the hardware, so you can see a camera here, it’s a RealSense camera, as well as a micro controller, so Intel NUC, and the Intel camera and the microcontroller were used to help detect up to 11 facial expressions.

Julie Shin Choi: So now the wheelchair user can operate and move using his own facial expressions. This is a whole new range of mobility that was unlocked because of AI. Such a powerful and memorable use case, and it’s just another example of the intersection of AI and human goodness. One more example from the field of healthcare, and I’m really passionate about healthcare and the AI applications that we’re seeing. This application that we see here is found… Again, it’s a collaboration between Intel and GE Healthcare, and in this case, what we see are deep learning algorithms that are inferenced at the edge in this powerful x-ray scanning machine. And the purpose here is to use AI and deep learning to identify cases of pneumothorax, or lung collapse, in record time. And the objective here is to augment physicians and to help prioritize cases so that doctors can get to people who are at higher risk faster than before. And this is really also helpful for parts of the world where doctors are scarce. So places like Asia and Africa, where the percentage of doctors is so low, and this type of AI can really help physicians get to patients much more quickly.

Julie Shin Choi: And one last example I want to bring up is the power and role that AI can have an accelerating diversity and inclusion. Last week, I had the privilege to go to North Carolina and attend an inclusion leadership summit that was organized by Lenovo and Intel’s chief diversity officers. And as we met, we brainstormed ways that AI could be used to eradicate bias in hiring practices, to accelerate ensuring that we have diverse and qualified candidates joining us and our organizations. We had a host of different chief diversity and inclusion officers in the room, as well as experts from law and policy and just AI research. So again, proving that when we bring disciplines together, we can really learn from one another to accelerate the kind of change that we all want to see at our companies.

Julie Shin Choi: So I want to kind of close with a summary slide on key takeaways and then we can have a conversation. In AI, good humans are needed because it’s such a powerful technology and it’s such an accelerant that really depends on algorithms at the heart, and these algorithms are coded based on assumptions that we make about data. So number one, we have to keep in mind, AI starts with data but ends with humans. It’s technology that’s being built for humans. So let us keep the end in mind as we design our AI products and solutions and keep the humans in the loop. Number two, I think it’s very important that we partner with people who really understand the human problems that we’re trying to solve. The Red Cross example, it couldn’t have been possible without the wealth of information that the Red Cross had, and it was truly a cross disciplinary effort. So we need to partner with domain experts.

Julie Shin Choi: Number three, be open-minded. AI is going to take a diversity of talents and tools. There’s really no one size fits all. We’re going to need CPUs, GPUs, FPGAs, these are all different kinds of hardware. Tiny edge processors. We’re going to need a host of different software tools. We’re going to need data scientists and social scientists, psychologists and physicists, marketers and coders to all work together to come up with solutions that are creative. It’s really going to take a village. And finally, let us be thoughtful. I know that in Silicon Valley people often say it’s important to go fast and to fail fast, but in AI, I don’t think so. I think we need to take time. We should be thoughtful and really, really careful and considerate about the assumptions we make as we create the tools that create the algorithms that feed the AIs. And certainly good humans will be needed every step of the way. So that is my last slide and I’m going to just now thank you all for listening and open up for questions.

Rachel Jones: Thank you so much, Julie. That was really fascinating. So yeah, everyone please send your questions. I’ve seen some people sending questions into the chat, but please make sure you’re putting them into the actual Q&A, that way people can upvote your questions and make sure that they get asked. So now our first question, is there any industry that you see where AI isn’t being used and what can humans do to bring AI into that industry?

Julie Shin Choi: Yeah, I mean that’s a great question. And honestly, we are seeing AI impacting virtually every industry that our customers are engaged in, from healthcare, to life sciences, to transportation, to retail, to finance, robotics, manufacturing. So most of those classic enterprise verticals are being transformed, are going through their AI transformations. What I will say is it’s still early days, even though it’s been about… I mean, I’ve been working in this space for five years. I always kind of mark that beginning… When you talk to researchers, they’ll say the beginning of AI really was deep learning, which really was 2012, but I kind of count from 2015, because that’s when Google really came out loud and proud as a machine learning company. So virtually every industry is being impacted by AI. Still early days. We’re about five years in and it’ll probably take the rest of, certainly my lifetime, the most of our lifetime, to kind of get to the maturity level that this technology is capable of.

Rachel Jones: Wow. So our next question, a concern that a lot of people have when they hear about AI is, “Oh, this is going to take all of our jobs and replace all the humans.” So what are your thoughts on that kind of anxiety?

Julie Shin Choi: I mean it’s very popular to say that, but I’m a firm believer that AI will not be replacing humans, it will be augmenting humans. So it’s helping us, not replacing us, because the whole… What we’re seeing, even in radiology, for example, radiology is a major transformation area that’s being transformed by AI faster than most because of the applicability of computer vision for x-ray imaging. But what we’re seeing is that physicians actually are welcoming the help of AI. It’s a great double check. When you have a 97% accurate algorithm that’s going to ensure that your patient gets the right diagnosis, even though the algorithm is sometimes even more accurate than you, especially if you’re tired, it’s an absolutely phenomenal double check, and so the end goal for the human in that case, in medicine, is to go and help that patient with the most accurate information that the human doctor has. So what we’re seeing is AI is truly helpful. It’s truly an augmenting type of technology and not a replacement.

Rachel Jones: All right. We just got another question. So this person says, “My daughter is still young, and if you had to mentor her so she’s prepared for the new AI world, what would you tell her?”

Julie Shin Choi: Yeah, that’s a great question. I have two children as well. I have 8 and 12. It’s funny, I will share an anecdote from dinner. A couple of months ago we were talking about the world and I have a junior high and an elementary, and the junior high, he said, “Well, I think that my generation is going to be spending most of its time solving the problems that your generation created.” And then my little one, who’s still elementary, chimed in right away, and he said, “With the help of our AI overlords, right?” These kids already, they’re so aware, and I think the advice to our children would be to really read books, play with one another, learn how to have friends from many different backgrounds, become the best humans they can be, because it’s not going to be robot overlords. We’re going to need good humans to program those AIs.

Rachel Jones: What’s the best way to learn AI?

Julie Shin Choi: Okay, so you guys probably have heard of Dr Andrew Ng and Coursera. Everything from on demand digital learning courses like the ones that Dr Andrew Ng pioneered, to tutorials on Intel’s AI website. There is so much knowledge out there right now around machine learning and deep learning that’s friendly for all levels and certainly Intel, we’re very committed to investing in preparing that kind of content and training. But I would encourage folks to check out all of those resources. We have certifications on our Intel developer community resources and we can connect you to those types of classes that take you step by step. Another partner organization that we like to work with on content delivery is O’Reilly Media. They have great courses online. A lot of these resources are free. I would say similar to the mobile revolution, when iOS and Android, all of those tutorials were popping up and hackathons every other day, we’re kind of seeing the same type of resources becoming available for AI and AI developers.

Rachel Jones: How pervasive is AI in the transportation industry?

Julie Shin Choi: Yeah, transportation is another really fascinating domain. Autonomous vehicles are a huge vertical being invested in. A lot of startup investment, a lot of institutional effort as well, so your established car companies and even airplane companies and shipping companies. We have a great use case from Rolls Royce that we’ve shared in the past. I didn’t realize that Rolls Royce also did transatlantic oceanic transportation autonomously, but they do, and it’s running on Intel. So transportation is going through a renaissance. It’s amazing. I think that actually–my husband works for an autonomous transportation startup, but again, early days. I always tell him, “You take that self-driving ride. For me, I think I’ll wait a little bit longer.” It’s still early days. A lot of innovation, a lot of promise and, yes, transportation is getting transformed.

Rachel Jones: So a big part of the AI conversation is about bias and how it can affect it. So what are your thoughts on that and how to limit bias in AI?

Julie Shin Choi: Yes, and bias is certainly a problem and it’s something that we, as a community of technologists and policymakers and social scientists, all different backgrounds, we need to attack this together. This was something that we discussed at the diversity and inclusion conference last week. A lot of it just comes down to let’s… There’s audits of algorithms. There’s ethics checklists, actually.

Julie Shin Choi: There are best practices that have been set up and I can actually introduce this community to our AI for Good leader, Anna Bethke, who is an expert in this domain and a wealth of knowledge. But we need to address bias with intentional and very purposeful conversations, because again, the algorithms are based on assumptions that humans code. So the only way that we can eradicate and deal with the bias issue is by talking to one another. The right experts in the room ensuring that have we checked that bias off the list? Don’t just assume that the coders know how to create a fair algorithm. I don’t think we can assume that. This is a very intentional action that we need to build into our AI development life cycles. The bias check.

Rachel Jones: All right. This is where we’re going to wrap it up. But Julie, thank you so much again. This was really great.

Julie Shin Choi: Okay. Thank you guys so much. Have a great day. Happy International Women’s Day.

Rachel Jones: Thank you. Happy International Women’s Day.