Alex Jackson has always been interested in the intersection between technology and ethics.
As a Carnegie Mellon University student working on particle physics research, she learned that the same data science methods she used there could also be used to collect data about people — and she knew she wanted to apply the same principles for “social good.”
Now, as a master’s degree student studying public policy and management with a focus on cybersecurity at CMU, she is launching a data hub to do just that. Part of a national organization called Data for Black Lives, the project aims to use numbers to tackle problems in Pittsburgh disproportionally affecting Black people and other people of color.
“As any city kind of rebrands and reinvents themselves, especially with a technology focus, it’s really important to ask who’s being brought in to that new economy and who’s being left out,” Ms. Jackson said.
“The new technologies we’re putting in the city … they’re not inclusive,” she said. “They’re not thinking about everyone.”
The data hub, which Ms. Jackson is working on with a team of 25 volunteers, is meant to “dig deeper into why these systems are failing.”
The team is focused specifically on problems addressed in a 2019 report from the University of Pittsburgh that found disparities for all women, especially Black women and girls, in the city.
Among other outcomes, that study found Black women are less likely to be employed and more likely to live in poverty than Black women in 85% of other U.S. cities that the study examined. Pittsburgh’s Black maternal mortality rate is higher than 97% of other cities.
The goal of the Data for Black Lives hub is to help amplify the work of existing organizations by providing them with data analysis in order to make recommendations for public policy, Ms. Jackson said.
For example, the group plans to look at mass incarceration in the greater Pittsburgh area and will analyze data from jails, including who is incarcerated, for how long, what services do they receive while incarcerated and what services do they not, and how has all that changed amid COVID-19.
She could not disclose what local organizations would collaborate with the Pittsburgh chapter but said the goal is to “put data analysis skills in the hands of everyday people who are doing really important critical community work that can transform whole cities.”
Pushing tech to change the status quo
Algorithms, machine learning and technology in general can be used to reinforce the status quo and, consequently, discriminatory practices already in place, Ms. Jackson said. She pointed to a debate about policing tools to be used in the city of Pittsburgh.
In September, Pittsburgh City Council passed a bill that would require the legislative body to give its approval prior to police use of facial recognition technology. Under the new law, council could grant approval if it finds the use of such technology doesn’t maintain or perpetuate bias and poses no risk to the civil rights and liberties of residents, but critics worry it is too lenient.
Facial recognition technology has been criticized for perpetuating discrimination and racism because some systems are not good at identifying Black faces, which could lead to unlawful arrests and falsely accusing people of crimes.
Predictive policing can also lead to an increased police presence in areas where crimes are not being committed, Ms. Jackson said.
“It puts people’s lives at risk when you’re talking about Black America in particular,” she said. “The algorithmic and technology issue is just another development in what has been a long, long history of discrimination in different sectors.”
Artificial intelligence and machine learning has progressed enough to the point where it can make a powerful impact on almost every sector in society, Ms. Jackson said.
“We’re finally at this point kind of where the promise of AI is coming to fruition … and we’re kind of at this moment where we get to decide what that looks like before it just completely takes on a life of its own,” she said. “Do we want to bake equitable outcomes into the system itself, or do we want these things to function as is and continue the status quo?”
In her view, academic circles and tech gurus are ready to have those conversations. The companies and researchers want their algorithms to be widely used, gain public trust and avoid intense scrutiny from third-party investigators.
Still, Ms. Jackson said, legislation and regulatory standards are needed in order to “be ahead of this issue.” That is one way the Data for Black Lives hub could come in, using data analysis to make suggestions about inclusive public policy.
Nationally, Data for Black Lives formed in 2016 as a nonprofit organization. So far, the Cambridge, Mass.-based group has worked to stop the expansion of a jail in New York City, encouraged cities to look at who is impacted by automated decision systems — or data-driven technologies that automate decisions about things such as how to allocate government resources — and called for states to share data of COVID-19 infections and deaths by race.
The group has also called on Facebook to join the Data for Black Lives movement as a way to promote equity on the platform and collaborate on data.
The Pittsburgh chapter, which does its data crunching from volunteers’ homes, plans to release its first data set next month.
At 23, Ms. Jackson said she often thinks about how much tech giants like Google and Facebook have changed in her lifetime — and how quickly they became a big part of people’s everyday lives.
While that might be a scary thought, to her it is also hopeful: If tech could change the world that quickly once, maybe it can happen again.
“It didn’t take that long for us to get to this point,” she said. “So the hope is it won’t take us too long to say, ‘Let’s put the brakes on some of this and critically examine what we’re doing and why we’re doing it.’”
“I believe Pittsburgh has a great opportunity to be at the forefront of thinking about truly equitable technology,” Ms. Jackson said. “We’re reinventing and trying to improve. It’s a great opportunity for us, and I believe we can get it right.”
Lauren Rosenblatt: lrosenblatt@post-gazette.com, 412-263-1565.
First Published: October 5, 2020, 10:00 a.m.
