Higher Education & ResearchIssue 164 records
Campus Operations
Campus-operations evidence on 2026-09-23 pairs Inside Higher Ed’s CAO survey—personal AI common, institutionwide operational transformation only 8%, facilities AI use 4%—with Jisc’s UK digital maturity assessment (sector 2.84/5) ranking AI capability among the lowest areas under funding and staff-capacity barriers. SIU System’s AI task force newly elevates multi-campus operations guardrails and Ellucian Smart Plan on Banner, while UCL Estates reports measured building-level energy cuts from BMS modernization and control tuning. Personal and building-level gains are advancing faster than independently verified institutionwide ops transformation.
- Evidence records
- 4
- Cross-source patterns
- 3
- Evidence classes
- 2 independent reporting1 public-sector association guidance1 vendor claim
- Outcomes
- 2 emerging1 mixed1 cautionary
- Source freshness
- 4 new this fortnight
- Research completed
- 2026-09-24
Choose a role to see its takeaway beside every record in the ledger.
Patterns across the evidence
Synthesis · Lighthouse Advisory interpretation
Personal AI productivity is widespread while institutional ops transformation and AI capability remain scarce
Inside Higher Ed’s CAO survey finds weekly personal AI use common but only 8% reporting institutionwide operational transformation and 10% a centralized strategy, while Jisc’s UK maturity assessment rates digital/AI capability lowest amid funding and staff-capacity barriers. Interpretation: campus-operations buyers should fund capability and workflow redesign, not equate leader copilots with estate-wide outcomes. Neither source measures the other’s geography.
Operating questionCan the institution show centralized AI ops ownership plus staff-capacity funding that move metrics beyond personal productivity into department workflow and facilities outcomes?
Supporting evidenceInside Higher Ed (with Hanover Research)Jisc
Measured facilities BMS optimisation advances while peer surveys show facilities AI use near zero
UCL Estates reports building-level energy cuts from BMS modernization and control tuning (including ~50% at Central House) as it enables analytics-led automation, while U.S. CAOs report only 4% facilities-management AI use and rare institutionwide ops transformation. Interpretation: estates OT/analytics programs may deliver operating benefits before “AI facilities” branding is common—claims must stay building-scoped and method-transparent. Sources are not causally linked.
Operating questionBefore marketing campus facilities AI, can estates show secure BMS baselines, named-building measured savings, and analytics-to-maintenance workflows?
Supporting evidenceUniversity College London EstatesInside Higher Ed (with Hanover Research)
Multi-campus ops AI frameworks are being drafted alongside SIS planning tools still awaiting usefulness proof
SIU’s task force elevates operations AI with legal/ethical/usefulness gates and describes Ellucian Smart Plan on Banner/Degree Works, while Jisc evidence shows capacity and fragmentation—not missing infrastructure—as the binding constraints on digital/AI maturity. Interpretation: system offices should sequence policy inventory and staff guidance before scaling SIS-linked planners. SIU has not published outcome metrics; Jisc is UK-sector level.
Operating questionDoes the system have an approved ops AI inventory and usefulness criteria before Banner-linked planning or paid Copilot expansion?
Supporting evidenceSouthern Illinois University System (reported by The Alestle)Jisc
Evidence ledger
Full record · every source keeps its link and limitations
Inside Higher Ed CAO survey: personal AI common, institutionwide ops transformation rare
Among 376 U.S. provosts, about 7 in 10 use AI weekly for personal work, but only 10% report a centralized AI strategy, 12% report department workflow optimization from institutional AI investment, 8% report institutionwide operational transformation, and only 4% report facilities-management AI use. Skills/capacity (51%) and staff resistance (50%) are the top limiters.
Why it matters, evidence and limitations
- Why it matters
- Direct campus-operations signal: personal productivity is not translating into measured institutional ops or facilities outcomes for most CAOs.
- Evidence and measured results
- Survey n=376, MoE 4.7 pp, fielded July–August 2026. Institutional uses: virtual assistants/chatbots 54%; research/data analysis ~40%; administrative scheduling 39% (up from 24%); facilities management 4%.
- Limitations and uncertainty
- Self-reported CAO perceptions; AI section sits alongside federal funding politics; facilities 4% is a usage share not an effectiveness measure; MoE applies to survey estimates.
Jisc: UK HE digital maturity averages 2.84/5; AI capability among lowest-rated areas
Eighty-three UK HE providers (88 responses; about half of UK universities) averaged 2.84/5 on Jisc’s digital maturity assessment. Reliable secure digital infrastructure scored highest; digital capability including AI capability scored lowest. Funding/investment was the most cited barrier, then staff capacity and time. Fragmented systems and uneven capability limit value from existing investment.
Why it matters, evidence and limitations
- Why it matters
- Campus operations AI programs inherit capacity and fragmentation constraints even where infrastructure is relatively strong—transferable caution for U.S. SLED estates and admin AI scale-ups.
- Evidence and measured results
- Assessment between May–June 2026 per companion report; 24 questions across six themes (culture, knowledge creation/development/management/exchange, digital and physical infrastructure). Sector score 2.84 on 1–5 reactive-to-integrated scale.
- Limitations and uncertainty
- UK funding context; aggregated sector score masks institutional variance; AI capability is named among lowest ratings but not published here as a standalone numeric subscale; self-assessment method.
SIU System AI task force sets operations goal including Ellucian Smart Plan on Banner
SIU’s AI task force defined four goals; goal three targets operational systems (grades, faculty reporting, scheduling) with legal/ethical/usefulness filters for staff tools across three campuses. SIUE Registrar describes Ellucian Smart Plan using Banner, scheduling, and Degree Works to rebuild semester plans and estimate added years/tuition when students reshuffle courses. Paid Copilot access is under consideration for some classes.
Why it matters, evidence and limitations
- Why it matters
- Concrete multi-campus campus-operations AI governance and SIS-adjacent planning tooling roadmap—still early, with student concerns recorded.
- Evidence and measured results
- On-record quotes from Mario Hayek, Tobias Merriman, Registrar Patrick Sears, and students; task force described as planning/research handoff rather than completed deployment metrics.
- Limitations and uncertainty
- Student newspaper interview; Smart Plan benefits described hypothetically (“in theory”); no measured ops KPIs yet; teaching/research goals dominate article length.
UCL Estates: BMS upgrade and building energy optimisation with measured building-level savings
UCL is upgrading Building Management Systems across an initial 35 buildings to replace obsolete, insecure controllers and enable analytics-led facilities management. Institution-reported results from control tuning and fault remediation include about 50% energy-use reduction at Central House, about 40% at Bentham House, and about 15% building energy-cost reduction (>£200,000/year) at Christopher Ingold via fume-hood setbacks. Narrative frames Building Energy Optimisation, Intelligence and Automation as the operating approach.
Why it matters, evidence and limitations
- Why it matters
- Facilities/OT modernization with measured building savings—contrasts with U.S. CAO survey showing facilities AI use near zero.
- Evidence and measured results
- First-party UCL Estates article dated 22 September 2026 quoting Assistant Director Chris Mitchell; named-building percentages and £200k figure without third-party audit in-page.
- Limitations and uncertainty
- First-party claims; savings attributed mainly to control improvement and fault repair, not independently validated generative AI; campus-wide extrapolation unsupported; comfort benefits narrative.
How to read this edition
Source findings, measured results and limitations come from the cited publications. Patterns, operating questions, role takeaways and implementation considerations are Lighthouse Advisory interpretation, stated as questions to validate locally rather than guaranteed outcomes. Vendor and operator claims are labeled as claims. Full research method.
- Independent reporting
- Independent reporting with attributable sources but without a formal evaluation design.
- Public-sector association guidance
- Practitioner guidance or an association-supplied case; not independent outcome evidence.
- Vendor claim
- A supplier-provided assertion that has not been upgraded to independent evidence.