Why AI matters at this scale

The SDSU Division of Research and Innovation (DRI) is the central engine for research, economic development, and sponsored projects at San Diego State University, a large public R2 institution. It supports thousands of faculty and student researchers across diverse disciplines, managing the complex lifecycle of research from ideation and grant procurement to compliance and commercialization. At this scale, with a vast research portfolio and administrative overhead, manual processes and data silos create significant friction, slowing the pace of discovery and innovation.

For an organization of this size and mission, AI is not a luxury but a strategic imperative to maintain competitiveness. It offers the ability to process information and administrative tasks at a scale impossible for human teams alone, directly impacting the university's research output, funding, and reputation. AI can automate low-value administrative work, freeing researchers and staff to focus on high-impact activities. More profoundly, it can augment human intellect, uncovering novel research pathways and collaborations hidden within the university's own data and the global corpus of knowledge.

Concrete AI Opportunities with ROI Framing

1. Automated Grant Intelligence & Development: The grant lifecycle is resource-intensive. An AI system trained on successful proposals and funder data can dramatically increase efficiency. It can scan funding opportunities, auto-populate boilerplate sections of proposals, and provide predictive scoring on a proposal's chance of success based on historical patterns. The ROI is direct: a potential 15-25% increase in award rates translates to millions in additional annual research funding, far outweighing implementation costs.

2. Cross-Disciplinary Research Discovery Platform: Research data is often trapped in disciplinary silos. A unified AI platform can ingest and semantically index research outputs—papers, datasets, patents, equipment—from across campus. It can then recommend unexpected collaborations (e.g., a computer scientist and a marine biologist) and identify nascent research trends. The ROI here is in quality and impact: fostering breakthrough, interdisciplinary work that attracts top talent, high-profile publications, and major grants.

3. AI-Enhanced Research Compliance & Reporting: Post-award management and compliance reporting are major administrative burdens. AI can monitor project spending against budgets in real time, flag discrepancies, and automatically generate progress reports by synthesizing data from lab systems and researcher updates. This reduces compliance risks and administrative FTEs, allowing staff to shift from reactive reporting to proactive support, improving researcher satisfaction and productivity.

Deployment Risks Specific to Large Public Institutions

Deploying AI at this scale within a large public university presents unique risks. Budgetary and Procurement Inertia is paramount; securing upfront investment for enterprise AI tools can be slow, competing with core academic needs. Data Governance and Silos are extreme, with sensitive human subject data, IP, and proprietary research scattered across decentralized colleges and labs, making unified AI training datasets difficult to assemble legally and technically. Cultural Adoption among tenured faculty can be challenging, as researchers may distrust "black-box" AI conclusions or perceive automation as a threat to academic rigor and independence. Finally, Talent Retention is a risk; building an in-house AI team is expensive, and the institution may struggle to compete with private-sector salaries for top AI engineers and data scientists, leading to capability gaps and project stalls.