“`html

Michael D. Smith and Klara Jelinkova at the IT Summit.
Campus & Community
IT Summit emphasizes balancing the challenges and benefits of AI
With technology firmly established, Michael Smith asserts that faculty and students must evolve into adept users
Investigating the essential function of technology in promoting Harvard’s mission and the promise of generative AI to transform the academic and operational landscapes were the primary subjects debated during the University’s 12th annual IT Summit. Organized by the CIO Council, the event on June 11 drew over 1,000 IT professionals from Harvard.
“Technology is foundational to every facet of Harvard,” stated Klara Jelinkova, vice president and chief information officer of the University, who commenced the event by commending IT personnel for their contributions across the institution.
This perspective was reiterated by keynote speaker Michael D. Smith, the John H. Finley Jr. Professor of Engineering and Applied Sciences and Distinguished Service Professor at Harvard University, who identified “people, physical environments, and digital technologies” as three fundamental pillars sustaining Harvard’s programs.
In his presentation, “You, Me, and ChatGPT: Insights and Predictions,” Smith examined the equilibrium between the challenges and prospects of employing generative AI tools. He pointed out the “explainability issue” inherent in generative AI tools, which may generate replies that seem plausible but lack clear reasoning: “Is this response accurate, or does it merely appear so?” Smith further emphasized user dissatisfaction stemming from poor prompts, “hallucinations,” and the danger of excessive dependence on AI for critical thinking, given its “readiness” to respond to inquiries.
While showcasing innovative coursework from students, Smith accentuated the transformative potential of “tutorbots,” or AI instruments designed with course content that can provide students with immediate, 24/7 support. AI is here to remain, Smith noted, urging educators to equip students for this future by helping them become adept and proficient users of the technology.
Asked by Jelinkova how IT personnel can assist students and faculty, Smith encouraged the audience to spot early technology adopters to “gain a clearer understanding of their objectives” and assist them through the “struggle” of mastering new tools. Comprehending these applications and nurturing collaboration can hasten adoption and “ultimately extend throughout the institution.”
The ethos of innovation and IT’s pivotal role at Harvard persisted throughout the day’s agenda, which was divided into four key themes:
- Teaching, Learning, and Research Technology included discussions where educators demonstrated their current experimentation with generative AI, including the Division of Continuing Education’s “Bot Club,” where instructors collaborate on AI-enhanced teaching methods and the implementation of custom GPTs and chatbots at Harvard Business School.
- Innovation and the Future of Services encompassed sessions on AI video experimentation, robotic process automation, ethical AI deployment, and a showcase of the institution’s newest AI Sandbox features.
- Infrastructure, Applications, and Operations offered an in-depth exploration of the remarkable efforts to bring the new David Rubenstein Treehouse conference center to fruition, involving the testing of new systems in a physical “sandbox” and the installation of thousands of feet of network cabling.
- Additionally, the Skills, Competencies, and Strategies breakout sessions examined the evolving skill requirements needed in contemporary IT—from automation design to vendor management—and discussed strategies for maintaining high-functioning, collaborative teams, including workforce flexibility and continuous development.
Amid the enthusiasm for innovation, the summit also considered the ecological impacts of new technologies. In a segment emphasizing Harvard’s leadership in IT sustainability—as part of its wider Sustainability Action Plan—presenters discussed how even minor individual efforts, like crafting more effective prompts, can significantly lessen the processing demands placed on AI systems. A panelist remarked, “Harvard has adopted AI, and with that comes the imperative to understand and thoughtfully evaluate its effects.”
“`