I recently had the privilege of attending and speaking at the Darwin Symposium, a yearly event hosted by the University of Delaware that brings together researchers exploring how artificial intelligence and high-performance computing are advancing work across a wide range of disciplines.
‍

Sitting through the presentations, what struck me most was that many of the discussions weren’t about AI itself, but about the real-world problems people are trying to solve using it. From disease prevention and coastal protection to higher crop yields and tools for the visually impaired, the work being highlighted was a reminder that the infrastructure powering AI often supports discoveries we don’t typically read about in the morning headlines, particularly in the technology sector.
‍

Several sessions highlighted how AI and advanced computing are being applied to some of the real, tangible challenges we face. A few examples that stood out included:

• Long COVID research. One presentation explored how researchers are using AI to analyze lipoprotein biomarkers in the liver in the hopes of enabling earlier identification of long COVID.
‍

• Agricultural resilience. Another project examined how AI-driven modeling could help improve lima bean yields, offering insight into how crops might perform under changing climate conditions.
‍

• Accessibility technologies. Researchers demonstrated vision-language models capable of translating images into descriptive text, creating tools that could help individuals who are blind or have low vision better interpret visual information.
‍

• Climate modeling. Another presentation examined how advanced computing can model how extreme weather may reshape America’s coastlines over time.
‍

• Education analytics. One project explored how data-driven tools can help high schools better understand their students' college readiness.
‍

Taken together, these examples shed light on how many of the most meaningful applications of AI are happening quietly within research environments. While consumer and business AI tools tend to dominate attention, it’s the work taking place in labs and computing clusters that may have a more far-reaching and lasting effect.
‍

Much of the public conversation around data centers focuses on their environmental and community impacts, including power demand, water usage, land use, and the strain large facilities can place on local neighborhoods. These are important discussions and deserve thoughtful engagement across the industry. As computing demand steadily increases, it’s reasonable that communities and policymakers need to better understand how data centers are built and how they affect the regions where they operate. It also presents an opportunity for data center developers to engage more meaningfully as long-term partners in those communities.
‍

Events like the Darwin Symposium underscore that infrastructure debate is only part of the bigger picture. Behind dialogue about power capacity and cooling systems are computing platforms that can expand what’s possible. They enable users to analyze massive datasets, model complex systems, and pursue discoveries that improve the environment, strengthen communities, and address wider societal challenges.
‍

Leaving the symposium, I found myself thinking about how easy it is to get caught up in the day-to-day conversations around infrastructure planning and design. We need to do a better job as an industry of creating mutual value between data centers and communities. Still, we also need these discussions to broaden the definition of impact. Events like Darwin help bring the wider context into view, revealing how the systems we design and implement ultimately support work that extends well beyond the infrastructure itself.

‍