Who benefits from artificial intelligence? This core question, which has been especially prominent during AI surges over the past few years, was front and center at a conference at MIT on Wednesday, as speakers and the audience grappled with the many dimensions of AI’s impact.
In one of the conference’s keynote talks, journalist Karen Hao ’15 called for a changed trajectory of AI development, including a move away from the massive transformation of data usage, data centers, and models being used to develop tools under the rubric of “artificial general intelligence.”
“This scale is unnecessary,” said Hao, who has become a prominent voice in AI discussions. “You don’t need this scale of AI and computation to realize the benefits.” Indeed, he adds, “If we really want AI to be broadly beneficial, we urgently need to move away from this approach.”
Hao is a former staff member wall street journal And MIT Technology ReviewAnd author of the 2025 book, “Empire of AI”. He has reported extensively on the development of the AI industry.
In his remarks, Hao outlined the staggering size of the datasets being used by the largest AI firms to develop large language models. He also emphasized some of the tradeoffs in this scale-up, such as the massive energy consumption and emissions of hyper-scale data centers, which also consume large amounts of water. Based on his own reporting, Hao also noted the human harm caused by global gig-economy workers manually inputting data for hyper-scale models.
Conversely, Hao offered, an alternative path to AI may exist in the example of AlphaFold, a Nobel Prize-winning tool used to identify protein structures. Hao said it represents “the concept of small, task-specific AI models that tackle a well-scoped problem that lends itself to the computational power of AI.”
He added: “It’s trained on highly curated data sets that relate only to the problem at hand: protein folding and amino acid sequences. … There’s no need for fast supercomputing because the datasets are small, the model is small, and it’s still unlocking huge benefits.”
In the second keynote address, scholar Paola Ricorte underlined the desirability of a purpose-driven AI approach, outlining several conceptual keys to evaluating the usefulness of AI.
“There’s no point in having technologies that won’t respond to the communities that are going to use them,” Ricorte said.
He is a professor at the Tecnológico de Monterrey in Mexico and a faculty associate at the Berkman Klein Center for Internet and Society at Harvard University. Ricorte has also served on expert committees such as the Global Partnership for AI, UNESCO’s AI Ethics Experts Without Borders, and the Women for Ethical AI project.
The event was hosted by the MIT Program in Women’s and Gender Studies. Manduhai Byandelgar, program director and professor of anthropology, provided introductory remarks.
The event, titled “Gender, Empire and AI: Symposium and Design Workshop”, was held in the conference space of the MIT Schwartzman College of Computing, with over 300 attendees for the keynote talks. A section of the program was also devoted to discussion groups, and there was also an afternoon session on design in half a dozen different subject areas.
In his speech, Hao decried the often vague nature of AI discourse, suggesting that it hinders more thoughtful discussion about the direction of the industry.
“Part of the challenge in talking about AI is the complete lack of specificity in the term ‘artificial intelligence,'” Hao said. “It’s like the word ‘transportation’. You can be referring to anything from a bicycle to a rocket. As a result,” he said, “when we talk about how to access its benefits, we have to be really very specific. What AI technologies are we talking about, and what AI technologies do we want more of?”
In his view, smaller-sized devices – similar to bicycles, by analogy – are more useful on an everyday basis. As another example, Hao mentioned Project Climate Change AI, which focuses on tools that can help improve the energy efficiency of buildings, track emissions, optimize supply chains, predict extreme weather, and more.
“This is the vision of AI that we should pursue,” Hao said.
Finally, Hao encouraged the audience to become active participants in AI-related discourse and projects, saying that the technology’s trajectory is not yet decided, and public intervention matters.
Quoting author Rebecca Solnit, Hao suggested to the audience that “hope establishes itself on the basis that we don’t know what will happen, and there is room to act in the immensity of uncertainty.” He also said, “Each of you has an active role in shaping technology developments.”
Similarly, Ricorte encouraged attendees to become active participants in AI matters, noting that technologies will work best when the everyday needs of all citizens are met.
“We have a responsibility to make hope possible,” Ricorte said.
