Humanities Artificial Intelligence Research Gets $11M Boost | HAVI 2025

International teams will develop AI tools to decode ancient texts, analyze film narratives, and map archaeological landscapes

Schmidt Sciences has announced $11 million in funding for 23 research teams working at the intersection of artificial intelligence and humanities scholarship. The awards, distributed through the Humanities and AI Virtual Institute (HAVI), support projects spanning archaeology, art history, literature, linguistics, and film studies across multiple continents.

The program takes a bilateral approach: teams will both apply AI to illuminate historical and cultural patterns while drawing on humanistic frameworks to shape more contextually aware AI systems. This reciprocal relationship distinguishes HAVI from traditional digital humanities initiatives, positioning scholars not merely as users of computational tools but as active participants in AI development.

Decoding Visual Language in Cinema

Among the funded projects, David Bamman’s Kinolab initiative addresses a fundamental gap in computational film analysis. While current AI models can process two-dimensional images, they cannot interpret why footage was filmed in particular ways or how technical choices create meaning.

Kinolab combines large-scale computational analysis with film scholarship across four areas: measuring the close-up’s psychological impact, classifying camera movements like dolly and crane shots, disentangling parallel storylines, and exploring relationships between visual and aural timing in television. Recent research in computational media studies has demonstrated AI’s capacity to reveal patterns across thousands of films, offering insights into representation, narrative structure, and aesthetic evolution that would be impossible through traditional frame-by-frame analysis.

Revealing Hidden Archaeological Landscapes

Jesse Casana’s team at Dartmouth College will create open-access tools for AI-driven archaeology, analyzing vast archives of satellite imagery and aerial photography to identify roadways, field systems, burial grounds, and settlement patterns. The project builds on decades of remote sensing research demonstrating that thermal imaging, multispectral analysis, and machine learning can detect subsurface structures and landscape modifications invisible to the naked eye.

Casana’s Spatial Archaeometry Lab has pioneered drone-based archaeological surveys, revealing extensive agricultural terraces and stone-built field systems beneath modern forests. The HAVI project scales this approach globally, potentially transforming how researchers understand ancient trade routes, environmental adaptations, and societal development across continents and millennia.

Technical Analysis Meets Art Historical Inquiry

Erma Hermens, director of Cambridge University’s Hamilton Kerr Institute, represents another dimension of AI humanities research. Her work on decorated medieval manuscripts exemplifies how AI can address challenges in scale, granularity, and stylistic intermixing that make large-scale art historical analysis difficult.

Technical art history, as outlined in recent scholarship, combines traditional connoisseurship with scientific methods from chemistry and physics. Hermens’ research on Dutch Golden Age flower paintings by Rachel Ruysch and her contemporaries uses computational imaging to analyze pigment compositions, brushwork patterns, and underdrawings. These investigations reveal workshop practices, attribution evidence, and artistic influences that reshape our understanding of 17th-century painting traditions.

The Stakes for Humanities Scholarship

Humanities research faces unique computational challenges. As Schmidt Sciences notes, current AI models train on massive datasets of contemporary information, modern languages, and two-dimensional media. Humanities scholarship, conversely, involves ancient or lesser-spoken languages, three-dimensional artifacts, varied material substrates, and relatively small amounts of ambiguous, culturally specific information.

“Rather than destroying the humanities, as many have feared, AI has a role in advancing the humanities, opening new avenues of scholarship,” said Brent Seales, the University of Kentucky professor who directs HAVI. Seales gained international recognition for developing “virtual unwrapping” technology that reads damaged ancient scrolls without physically opening them.

Research examining AI adoption in digital humanities reveals growing integration of machine learning for text analysis, image recognition, and pattern detection. Studies document successful applications ranging from authorship attribution and sentiment analysis of historical documents to automated transcription of handwritten manuscripts and reconstruction of fragmentary texts.

A Virtual Institute Model

The virtual institute structure allows geographically distributed teams to collaborate without requiring physical proximity. This proves particularly valuable for humanities research requiring access to dispersed archives, multilingual expertise, and cross-cultural perspectives.

The 2025 cohort joins two inaugural HAVI awards granted earlier: Sorbonne University’s Digital Delacroix project analyzing 19th-century French Romantic painting, and EduceLab, a heritage science facility applying AI and micro-CT imaging to cultural artifacts.

Schmidt Sciences, founded in 2024 by Eric and Wendy Schmidt, focuses on accelerating scientific breakthroughs through advanced computational tools. Beyond HAVI, the organization supports research in AI and advanced computing, astrophysics, biosciences, climate science, and space exploration.

“Our newest technologies may shed light on our oldest truths, on all that makes us human,” said Wendy Schmidt, co-founder. “HAVI is poised to change not only the course of scholarship, but also the way we see ourselves and our role in the world.”

The complete list of 2025 HAVI projects is available on the Schmidt Sciences website. Applications for the next funding cycle are due March 13, 2026.

---

ABOUT THE ORGANIZATIONS