In the age of artificial intelligence-generated artwork and ChatGPT-ed assignments, AI is often criticized for reducing the need for human work. Yet, Northeastern’s Civic A.I. Lab, directed by Saiph Savage, assistant professor in the Khoury College of Computer Sciences, is using AI to make it easier for people to find and keep fulfilling jobs.

The lab, which is housed in the Khoury College of Computer Sciences, tackles issues related to gig work — temporary work performed by an independent contractor — and political disinformation by creating computational systems that take human needs into account. 

Aishwarya Abbimutt Nagendra Kumar, a research assistant who is pursuing a master’s degree in data science at Northeastern, focuses on helping freelance workers find gigs. Her projects use AI technology to create programs that allow workers to better understand the job market and progress in their career paths. 

The system she developed consists of the “Market Insights Engine,” which analyzes in-demand skills and trending careers, and the “Personalized Career Adviser,” which uses individuals’ transferable skills and goals to create personalized career recommendations. 

“[The tool shows] what the market trends are right now and what careers would make you money,” Kumar said. 

While this technology is not available to the public yet, Kumar hopes that it will eventually be free to use, allowing workers to develop useful skills and grow financially. 

The lab also works to make freelance employment experiences more sustainable and safe. Maya De Los Santos, a former research assistant at the Civic A.I. Lab and fifth-year electrical and computer engineering combined major, initially focused on interviewing female Latin American and Caribbean crowd workers to understand their concerns and challenges. Crowd work refers to work in which gig workers are hired temporarily to complete tasks for employers but are not necessarily entitled to being paid for the tasks they complete. 

“Education was very important to them because they often had to teach themselves how to complete their tasks because their employers would pay them based on their output and not all the time and work it took them to deliver that task,” De Los Santos said. 

Lack of community among crowd workers was another prevalent problem identified by the women De Los Santos interviewed.

“They had no way of speaking to one another or sharing their knowledge or experiences or concerns with one another because the platforms and the employers purposely try to keep them separate from one another,” De Los Santos said. 

Using this information, she developed a chatbot that allows women to build connections and learn skills from one another. The chatbot has multilingual capabilities to accommodate the diverse needs of the Latin American and Caribbean communities. 

De Los Santos also collaborated with employees worldwide to develop sousveillance systems designed specifically to assist gig workers. Sousveillance refers to the act of employees monitoring their workplace or employers, as opposed to surveillance, which involves employers tracking their workers. Together, they conceptualized a well-being tracker that reminds workers to periodically stretch or rehydrate. This is especially relevant for those who must work odd hours to accommodate their employers who live in different time zones. De Los Santos also developed an anonymous client review board that allows workers to see which clients are likely to be safe to work with.

Viraj Upadhyay, a recent graduate of the human factors master’s program at Northeastern and research assistant in the Civic A.I. Lab, works to ensure job security for journalists, specifically those who are targeted by political entities who aim to portray them in a negative light. 

“I’m primarily working on building visualization tools for journalists that use AI to perform analysis to find and underline patterns in the data, and also topic modeling, sentiment analysis and analyzing the content as well as how the comment section or the audience interacts with it,” Upadhyay said. 

He uses AI systems to analyze long pieces of video media, like YouTube videos or news segments, to identify the context in which journalists’ names are mentioned and the sentiments associated with them. Upadhyay also uses this technology to quickly identify the video’s topic and visualize it into a word cloud summary. The tool also offers a text-to-speech feature to ensure accessibility. These systems allow journalists to quickly detect when they are being targeted.

Upadhyay’s work also centers on limiting political disinformation, using AI to source image content, verify the validity of information and identify the source of information that is being spread. 

Despite the complexity of these technologies, Upadhyay said he values straight-forwardness over intricate solutions. 

“I am interested in providing design solutions, finding out how we can solve a particular challenge as simply as possible and not bringing too much technology into it, but trying to solve it as efficiently as I can,” he said. 

Ultimately, the Civic A.I. Lab and its contributors work to improve the accessibility of AI and ensure that it is used to benefit workers — not take from them. 

“I found [Savage’s work] very impressive because one thing I have always thought about when using … AI was how to use it to make an impact and use it for something positive,” Kumar said.