Audrey Lorvo, a senior at MIT, is deeply engaged in the important field of AI safety, which focuses on making advanced AI models reliable and beneficial for humanity. As AI technologies become more sophisticated, key concerns arise around technical aspects like robustness and alignment with human values, coupled with societal issues such as transparency and accountability. The field also addresses the existential risks that powerful AI systems may pose.
“It’s crucial to ensure AI is developed responsibly, especially as we edge closer to achieving artificial general intelligence (AGI),” explains Lorvo, who studies computer science, economics, and data science. AGI refers to AI with the ability to match or surpass human cognitive performance.
As a Social and Ethical Responsibilities of Computing (SERC) scholar at MIT’s Schwarzman College of Computing, Lorvo examines the implications of AI in automating its own research and development. A member of the Big Data research group, she researches the social and economic impacts of AI’s capacity to enhance its own development while also exploring effective ways to communicate these findings to various audiences, including legislators and strategic decision-makers.
Lorvo emphasizes the necessity of critically evaluating the rapid progression of AI technologies, urging organizations to establish frameworks to address associated risks. “We must ensure that humanity benefits from AI, while maintaining control over its development,” she asserts. “It’s vital to prioritize safe development practices.”
The Value of Human Engagement
Upon joining MIT, Lorvo was determined to pursue studies that bridged the gap between science and the humanities. The diverse options available at MIT made her decision challenging.
“There are countless ways to enhance community well-being,” she notes, “and MIT provides numerous avenues for exploration.”
Initially drawn to economics for its emphasis on measurable impacts, Lorvo also explored mathematics, political science, and urban planning before settling on Course 6-14.
“Professor Joshua Angrist’s econometrics courses highlighted the significance of economics for me, while the data science and computer science components intrigued me because of AI’s expanding potential,” she explains. “We can leverage these capabilities to tackle some of society’s most urgent issues.”
In addition to her major, Lorvo is pursuing concentrations in urban studies and planning as well as international development.
Through her narrowing focus, Lorvo has found common ground with fellow members of the MIT community, such as those in the MIT AI Alignment group, where she deepened her understanding of AI safety. “Students are genuinely invested in making a marginal impact,” she states.
Marginal impact refers to the additional benefit derived from a specific investment of time, effort, or resources, rather than the total outcome. This approach aids individuals in determining where to concentrate their efforts, a strategy that appeals to Lorvo.
“In a world with finite resources, applying a data-driven method to address our most significant challenges can maximize the good accomplished,” she asserts. “Considering the marginal impact of our career choices can provide valuable insights into achieving social change.”
Lorvo also values MIT’s holistic approach to education, which has allowed her to explore subjects like philosophy through MIT Concourse, a program encouraging dialogue between science and humanities disciplines. Concourse aims to assist participants in finding direction and purpose in their scientific, technical, and humanistic endeavors.
Student Experiences at the Institute
Outside the classroom, Lorvo dedicates her time to creating memorable experiences and building connections with her peers. “I’m grateful that I can balance coursework, research, and club activities with personal interests like weightlifting and community initiatives,” she notes. “MIT has an abundance of clubs and events to engage with.”
These experiences have broadened her perspective and introduced her to new interests, shaping her life and career path positively. Fluent in French, English, Spanish, and Portuguese, Lorvo appreciates the international opportunities available at MIT.
“I’ve interned in Santiago, Chile, and Paris through MISTI, where I helped test a water vapor condensing chamber designed in collaboration with the Madagascar Polytechnic School and the Tatirano NGO,” she shares. “I’ve also learned about addressing economic disparities through my classes in International Development and D-Lab.”
As president of the Undergraduate Economics Association at MIT, Lorvo connects with like-minded students while enhancing her understanding of economics. She enjoys the relationships she’s forming and actively participates in various events organized by the association. “Even in my senior year, I’ve discovered new communities to engage with,” she remarks. “I encourage other students to pursue groups and classes that ignite their passions throughout their time at MIT.”
After graduation, Lorvo aims to continue her exploration of AI safety and the governance strategies that will ensure the safe and effective deployment of AI technologies.
“Good governance is vital to the development of AI, ensuring that humanity capitalizes on its transformative possibilities,” she emphasizes. “We must continuously monitor AI’s advancements as the technology progresses.”
With an eye toward understanding technology’s potential impact on society, promoting positivity, and nurturing environments where innovative ideas can flourish, Lorvo is motivated in her efforts. “I have always aspired to contribute to enhancing people’s lives, and I view AI as both our greatest opportunity and challenge,” she explains. “The AI safety field can greatly benefit from individuals with interdisciplinary backgrounds, and I encourage anyone passionate about shaping our future to delve into this fascinating area.”
Photo credit & article inspired by: Massachusetts Institute of Technology
