Colleges and universities continuously adopt new technologies to improve teaching, learning and campus operations. As Artificial Intelligence (AI) use grows, higher education institutions must modernize infrastructure, protect sensitive data and prepare students for an evolving workforce. To successfully navigate these changes, key priorities include responsible AI adoption, secure infrastructure, data governance and ongoing faculty and student training.
Carahsoft, The Trusted Education IT Solutions Provider™, held the annual EdTech Talks Summit where education and IT industry experts explored how AI, cybersecurity and workforce development are shaping the future of higher education and creating new opportunities for innovation. Below are five insights that emerged from these discussions.
AI Readiness Begins with Connected Infrastructure
Agentic AI depends on underlying systems. Neal Tilley, Cisco Education Business Development Manager in the U.S. Public Sector, noted that evolving AI strategies, legacy technology and data silos complicate adoption. He encouraged technology partners to offer high-level plans that coordinate resources and balance cloud and infrastructure priorities.
Splunk speakers reinforced that AI depends on strong infrastructure. Visibility should precede AI agents, which can then operate in a well-understood environment. Hewlett Packard Enterprise (HPE) added that AI-ready systems need robust security, speed, broad distribution and firewalls to enable safe agentic AI experimentation. In one U.S. university, a software-defined infrastructure platform reduced graphics processing unit (GPU) per-hour costs, accelerated researcher access to GPUs and connected applications across campuses.
Before expanding AI initiatives, institutions should ask:
- Can existing infrastructure support emerging AI applications?
- How is sensitive institutional data being protected?
- Are clear policies in place for responsible AI use?
- How can AI reduce administrative workloads and improve the student experience?
These questions help universities move beyond isolated AI experiments toward practical applications that support faculty, streamline operations and improve learning environments.
Responsible AI Governance Is Built Together
Panelists identified data privacy as the top priority. Tilley urged universities to develop AI policies with stakeholders, using National Institute of Standards and Technology (NIST) responsible AI requirements as a reference. Splunk recommended starting policy conversations with cybersecurity requirements and keeping AI simple enough to save faculty time. HPE emphasized that AI should handle “side-rule” work so people can focus on creativity.
Nutanix outlined three pillars of sustainable AI governance:
- Control: Tool-level and token controls, central application programming interface (API) security and agent audit tracking
- Choice: Flexibility across on-premises, cloud and virtual deployments and models
- Sustainability: Unified policies, visible inventories, cost tracking per outcome and avoiding irreversible decisions
A live Superhuman Go demonstration showed distinct student and faculty agent controls, including agents that build Canvas course pages and recommend improvements to student work using course rubrics.
Identity and Zero Trust Anchor Campus Cybersecurity
Cybersecurity begins with data ownership. Shawn Tieman, Product Leader in the Enterprise Security Group at Broadcom, explained that most attacks identities rather than software vulnerabilities. Understanding exposure across managed and unmanaged devices and Software as a Service (SaaS) is the first step.
NinjaOne highlighted education’s diverse users and devices, requiring support solutions both on and off premises. Building Zero Trust into daily operations, automating remediation and meeting standards such as FedRAMP and System and Organization Controls (SOC) 2 keep systems current while preserving academic freedom and continuity.
Mathew Woodyard, Director of Threat Intelligence at Okta, described how large language models (LLMs) accelerate and scale phishing, including voice phishing and specific institution kits. Mandatory phishing-resistant authenticator enrollment offers strong protection. Co-managed models between managed service providers (MSPs) and universities, supported by security operations centers (SOCs), show real promise.
Zero Trust and clear data governance policies reduce risk while maintaining reliable access to educational resources. Faculty and staff also need cybersecurity awareness and training to protect sensitive information when using AI applications.
Securing the AI Estate Requires Continuous Visibility
David Klein, Solutions Consultant Manager at Palo Alto Networks, explained that generative AI (GenAI) is non-deterministic, unlike traditional applications, so security teams benefit from live monitoring grounded in the NIST AI Risk Management Framework (RMF). Institutions should map data flows, intended impacts and AI use boundaries. Key practices include:
- Zero retention clauses in AI provider contracts
- Continuous security testing and aggressive red-team exercises
- AI gateways with inline data loss prevention (DLP) to block risky prompts before they reach models
- Hard fail-safes and Zero Trust tool access controls as agents gain API access
Visibility also protects admissions. Tyler Merwin, Director of Sales at AMSimpkins & Associates (AMSA), warned that AI-generated content and voice cloning can impersonate applicants. Internet Protocol (IP) addresses, virtual private networks (VPNs) and unusual activity prompt further review but do not independently prove fraud.
Workforce Readiness Grows Through Gradual Adoption
As AI transforms the workplace, colleges and universities must prepare students and educators to navigate emerging technologies. While AI adoption is increasing, many institutions are still determining how to integrate it into academic environments while addressing concerns about training, privacy and responsible use.
Mark Henderson, Vice Chancellor and CIO at the University of Pittsburgh, observed that institutions differ in AI readiness and have no universal timeline. He recommended gradual introduction, giving faculty and staff time and resources to adapt. Secure AI environments, professional development and clear guidelines increase confidence in using these technologies.
Brian Dreyer, Senior Director of Product at Cloudforce, added that AI access is outpacing formal support structures. Institutionally supported tools, clear policies and accessible training build confidence and consistency. Leaders should ask:
- How prepared is the institution to support AI adoption?
- Where are faculty, staff and students using AI today?
- What is the biggest barrier to responsible adoption?
Institutions must equip faculty to teach responsible AI use and help students develop AI literacy, practical digital skills and critical thinking for an increasingly AI-driven workforce.
Looking Ahead
Carahsoft’s annual EdTech Talks Summit highlighted how collaboration between higher education institutions and the IT industry supports responsible technology adoption and secure, innovative, accessible learning environments for students and educators. From modernizing AI infrastructure and strengthening cybersecurity to preparing students for the future workforce, emerging technologies are reshaping the higher education landscape. As digital transformation continues, institutions must balance innovation with security, governance and training.
To learn more about utilizing IT for education initiatives, explore Carahsoft’s Education Technology portfolio.
Connect with our team today at EDUmarketing@carahsoft.com or (571) 591-6070 to begin your journey towards a more efficient educational environment.
