TEMPE, Ariz. — Robots are taking our jobs!
Human fear of artificial intelligence has long been a polarizing topic in American pop and tech culture, politics, and even some academic circles, and it’s only getting more so as AI technology advances.
AI technology has evolved significantly since its inception in the 1950s. An Arizona company is doing its part to dispel common AI misconceptions by educating the next generation of AI engineers.
“I think we can do a lot of mystical things with it, like some people call AI a black box, and it’s such a catch-all term because it’s complex, it’s confusing, and it’s deep in its analytical nature,” assistant Alec Evans said. Director of Data Science for DriveTime, an online used-car dealership and finance company based in Tempe.
In October and November, DriveTime partnered with the Mark Cuban Foundation to host Arizona’s first Mark Cuban Foundation AI Boot Camp for high school students.
“And I think that demystifying what we’re doing here with students and saying, ‘You use it every day, you see it every day, you don’t know what it is,’ can really open students up.” This new career Minds on occasion,” Evans said.
Your browser does not support HTML video.
AI is a blanket term covering many innovations such as machine learning and natural language processing. However, AI still has limitations, which make it unable to perform many tasks that humans routinely perform.
“The biggest problem comes when there’s a mismatch between the actual technology’s capabilities and society’s perception. It’s hard to reconcile them, and that creates all kinds of problems,” said Aviral Srivastava, a professor at Arizona State University’s School of Computing and Augmented Intelligence.
Society is already dependent on AI technology. From virtual assistants like Siri, Google and Alexa to apps like Netflix and YouTube that use machine learning algorithms to recommend content tailored specifically to your interests, AI is part of our everyday lives.
However, most AIs employed by companies go unnoticed: in banking apps, email and spam filters, auto-correction tools, personalized ads and thumbnails of videos viewed on YouTube.
Chatbots use natural language processing to determine and interpret what the user is asking and then schedule the correct response according to the language it has already processed. Autonomous driving has also made progress.
AI as a career path
DriveTime has been using AI tools and analytics to help ease the customer’s used vehicle buying process, and Evans said they are at the heart of what differentiates the company from competitors.
“You’ve got people looking for cars first, and you’ve got people looking for value first,” he said. “What we want to make sure is that our experience can accommodate both of those shopping experiences.”
Students from across the Phoenix metro were selected for Boot Camp, a four-week development program that taught them AI, machine learning, natural language processing and more. Mark Cuban is a billionaire entrepreneur and owner of the Dallas Mavericks, and his foundation focuses on providing high school students with opportunities to learn AI and network in the tech space.
The boot camps began in 2019, and this year will reach more than 550 students in 26 US regions — 20 in the Valley — and are hosted by companies using AI technology. According to the foundation’s website, the camps allow students to meet employers in the corporate world and connect with adults who have jobs in STEM, AI and machine learning. Students do not need to have coding experience.
Katica Calderon’s father signed the 14-year-old up for a boot camp based on her interest in AI and how it can help others.
“All the knowledge I have about AI is from my father,” said Calderon, who hopes to one day work in the field. “He gave me Alexa, so I started researching how it worked and I got really interested in that kind of stuff.”
Building networking skills and helping students understand that professions using AI technology offer practical career paths is what makes the boot camp important, said Drivetime’s Robin Jordan, whose title is Head of People.
“It’s kind of this new career journey, and it’s where the future is going,” Jordan said, noting that companies must embrace the technology, and start educating about it early. “I think as people start to understand what it can do, and not be afraid of it and embrace it, I think you’re going to see a lot more people trying to figure out how to use it better.”
Robots and their work
Some worry that AI tech will make some human jobs obsolete.
Human employees can handle multiple tasks at once and think critically about what they are doing and why. Although machines cannot think, they can perform mundane, single-minded tasks faster and more efficiently than humans.
“Even when people do (low-stakes tasks), they’re not very good at it,” Srivastava said. “So they also make mistakes, and AI can do a good enough job, and in fact, it can do a very good job because it can look at all the parameters systematically and look at a much larger amount of data than humans. Ever have the patience to look.”
Aside from human error and fatigue, another important difference between machine learning algorithms and humans is that humans are biased and machines are not. Algorithms don’t create trends or spot anomalies in data, they just calculate.
“The way (machine learning) algorithms are designed, they’re only reflecting what the data is telling them. So if your data is biased, your predictions are biased,” Srivastava said. “Because algorithms are so efficient, we’re starting to use them in places where fairness is an issue. Before, algorithm standards were never an issue — why? Because they were never used in these situations.”
If a company uses an algorithm to determine how much to compensate employees, the algorithm will rely on previous compensation data, Srivastava explained. But if an employer isn’t paying a certain employee or group a fair wage, AI will continue the negative trend of compensation, seemingly reaffirming human bias.
Earlier this year, the Forbes Technology Council, an organization of technology executives, compiled a list of 15 jobs and functions they expect to be automated within a decade — including data coordination, accounting, driving, simple customer service, sales and marketing. That all elements are at risk.
Although an algorithm can relieve one person of routine work, that algorithm requires at least one person to create and shape it. The uptake and expansion of AI tech doesn’t mean jobs will be eliminated, but they can be repurposed into better-paying tech jobs.
Evans shared the story of her grandmother saying she never wanted a robot to drive her car.
“But, Grandma, what if you find out that people like me have taught robots to drive? I’m the one who works on those algorithms and teaches them what to do and what not to do.
Stay in touch with us anytime, anywhere