There’s hardly a single piece of software we use in our day-to-day lives that hasn’t had some sort of AI agent injected into it. But as ubiquitous as artificial intelligence is in the current landscape, some of its characteristics are hidden from view. Society is in the midst of a massive technological shift, and as we contend with this new technology that is pushing so hard to assert itself on all of us, journalist Karen Hao is ready to expose the behind-the-scenes risks and corruptions that come with it.
This past Wednesday, Hao gave a talk in the Dodd Center related to her recent book, “Empire of AI.” As a journalist with experience covering AI as well as experience training other journalists on how to cover it, Hao provided an in-depth look into the many ways that AI is connected to structures of power and continues to perpetuate them.
One of the stand-out aspects of Hao’s lecture was how she illustrated the sheer variety of ways in which AI impacts people all over the world. According to Hao, you may be one of the many for whom AI poses an imminent threat, feeling the brunt of the pressure to adapt to the menacing shadow cast by AI on the job market. But beyond these changes that are much more immediate to Connecticut residents, Hao brought to light the myriad of other ways that AI is impacting lives globally.
For AI to be a commercially viable product, large amounts of invisible labor is necessary to filter out the disturbing content that these LLMs can generate. For the hundreds of content moderators sifting through this disturbing material, this work takes a sizable toll on their mental health, Hao said.
“They were going through text… that represented the worst things on the internet,” Hao said. “That work ultimately paid… barely anything, on average $1.46 and $3.74 an hour.”
This exploitation on the part of AI companies, which Hao argues mirrors more traditional imperialism, is the main focus of her book.
“We need to stop thinking of companies like Open AI as merely businesses providing products and services,” Hao said. “They’re new forms of empire that are consolidating a historic amount of economic and political power, terraforming our earth, reshaping our geopolitics, upending our educations, our social lives, and your future careers.”
Many in the audience shared a sense of concern about the consequences of AI, as well as some sense of hope about our ability to adapt to it in the future.
“I’d like to believe that I don’t interact with AI a lot. I don’t really use any generative AI platforms like ChatGPT,” said Jason Wong, a second-semester undecided student. “I feel like the most I interact with AI, especially generative AI, is when you do Google searches and the automatic, generative AI response comes up to summarize everything about the topic you search for.”
Other attendees have had to face these AI implements in similar ways. History professor Melanie Newport stated that “I find myself having to find ways to work around the Microsoft chatbot in order to access my word documents.”
But for her, the UConn community is working to adapt.
“At the department level, we have been having meetings to discuss how to deal with AI in the classroom.” Newport said. “I think it’s really hard for people who are trying to teach writing and research. I know for myself… we just do everything in-person in class, which I think is actually fun. We just have a really human experience together.”
Professor Martha Cutter also talked about the significance of humanity in the classroom. “I’m an English professor… I’m going to hear your voice. For the first draft… I don’t care about whether there’s grammar problems. I want to hear you speak.”
Beyond just this question of academic honesty within the classroom, it’s also important to consider how AI intersects with the job market, and to what extent students should be prepared to use it in the workforce.
“I come in from an engineering standpoint,” said Daniel Burkey, Associate Dean for Undergraduate Education & Diversity at the College of Engineering. “An engineering standpoint is sort of technology and workforce focused. And so there’s been a lot of discussion at the university… What are employers who are going to hire our engineering graduates expecting their students to know? When students leave here, what is the next thing after here? When do you expect them to know about it? And what is our responsibility as the university to provide that?”
AI has presented a fundamental shift in the way that education works, and as the university community continues to navigate it, many different strategies are coming to light. Beyond just the fundamental question of academic honesty, artificial intelligence is pushing both professors and students to protect human connection in the classroom and raising questions about how to balance both learning and preparedness for the workforce.
As these new AI empires continue to build themselves off the backs of common people, as society continues to wrestle with the economic, environmental and social fallout, as we approach newfound questions about information and privacy, Hao emphasizes the importance of community connection and mobilization in preserving human rights. Through our collective voice, we can assert ourselves and negotiate a positive future that balances both human rights and technological development, she said.