From the College Athletics edition of September 15, 2026
ASU football projects prep-time savings from an on-device play-card AI agent
Intel / Arizona State University Athletics · Higher education athletics / semiconductor vendor · United States (Arizona; Division I football)
- Publisher
- Intel Newsroom
- Original publication
- 2026-09-04
- Source retrieved
- 2026-09-15
What happened
Intel describes a six-month ASU football proof-of-concept: a local AI agent on Lenovo/Intel vPro laptops that maps opponent film to digital play-card diagrams, with coaching staff expecting up to 15–20% less game-prep time and on-device handling of playbooks, health metrics, and film.
Why it matters
Why now: first major attributed U.S. Division I coaching-workflow agent claim found in this research window that was not yet in the college-athletics archive, despite publication shortly before the 2026-09-13 edition. Speaks directly to coaching operations and local-data architecture for athletic departments.
Evidence and measured results
Newsroom bullets and body state Intel–ASU collaboration, custom play-card agent behavior (upload film → identify positions/movement → whiteboard drawings), projected 15–20% prep-time reduction, and on-device IP/privacy framing. Named ASU athletics staff quote on efficiency. No independent measurement table is provided.
Limitations and uncertainty
Projected—not measured—savings; vendor authorship; no published accuracy of formation/route extraction; not evidence of wins, injury reduction, or transferable ROI for smaller programs.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-09-15; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Coaching staff spend many hours manually converting opponent film into practice play cards under travel and data-sensitivity constraints.
- Stakeholders
- Head coach / coordinators
- Football operations
- Campus IT / information security
- Athletics compliance
- Discovery questions
- How many coach-hours per opponent currently go to play-card creation?
- May game film or playbooks leave university-controlled systems?
- Who reviews AI-generated diagrams before they enter practice?
- What error rate in formation or assignment diagrams would halt the pilot?
- Value hypothesis
- A managed on-device film-to-diagram assistant may reduce manual drawing time if accuracy and review burden are acceptable—without claiming measured ASU savings as guaranteed.
- Potential engagement
- 4–6 week play-card pilot on managed laptops with timed baseline vs assisted workflows and an explicit stop rule for diagram error rates.
- Unsupported claims
- Proven 15–20% prep-time savings at the prospect’s program
- On-field competitive advantage or win-rate improvement
- Automatic NCAA or privacy compliance from on-device processing alone
Pre-sales engineering
Role takeaway
- Fit
- Endpoint agentic workflow for film understanding and diagram generation; not a substitute for roster, medical, or recruiting systems.
- Architecture and integration
- Managed Windows endpoints; local model/runtime; controlled film import; optional sync only of approved artifacts to university storage—not raw biometric streams.
- Prerequisites
- Endpoint management and disk encryption
- Labeled film/diagram pairs for acceptance testing
- Coach review UI or checklist
- Clear data-classification for film vs health metrics
- Constraints
- Vendor-reported ASU results are projections
- GPU/NPU capacity and battery limits on travel days
- Model drift across conferences and camera angles
- Security
- Keep proprietary film on-device or in university vaults; restrict USB/cloud exfil; separate athlete medical data from coaching film workspaces.
- Proof of value
- Prep minutes per opponent with vs without the agent
- Coach-rated diagram correction rate
- Count of unintended data egress events during the pilot
Delivery
Role takeaway
- Work
- Stand up a small coach cohort, configure devices, define diagram QA, capture time-motion baseline, and run assisted prep for a fixed slate of opponents.
- Dependencies
- Film library access
- IT imaging/vPro management
- Coach availability in-season
- Security review of local AI stack
- Ownership
- Football operations owns workflow acceptance; campus IT owns device security; analytics staff (if any) owns measurement.
- Skills and adoption
- Short coach training on when to trust, edit, or discard diagrams; escalate ambiguous formations to human film study.
- Governance checkpoints
- Pre-pilot data-classification approval
- Mid-pilot error-rate review
- Go/no-go before expanding beyond football
- Acceptance criteria
- Documented reduction in prep minutes without rise in diagram corrections beyond agreed threshold
- No unauthorized film upload to external AI services
- Written owner for each practice-plan change
- Risks
- Hallucinated alignments entering practice
- Over-reliance during compressed road weeks
- Vendor narrative outrunning local measurement
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Prefer on-device or campus-controlled inference for proprietary film and playbooks when cloud upload is unacceptable; still require device management (vPro-class controls), model update channels, and offline evaluation of extraction accuracy before scaling beyond a pilot cohort of coaches.
Governance
Who approves, reviews and stays accountable for outcomes?
Treat vendor projected time savings as a hypothesis for a bounded pilot with a pre-registered baseline of prep hours, error review of generated cards, and clear human ownership of play-call and teaching decisions.
Security and privacy
What data, permissions and controls need testing?
On-device processing can reduce third-party film exposure, but laptops remain high-value targets; protect lost/stolen devices, control who can export diagrams, and separate any athlete health metrics from general coaching accounts.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
May reduce repetitive diagramming load for staff but can deskill film-study practice if juniors only accept AI output; train multiple staff on review so knowledge is not trapped with one power user.
Procurement
What should contracts, pricing and exit terms secure?
Bundle hardware, management licenses, and model support; require acceptance tests for diagram accuracy and data residency in the contract rather than relying on newsroom percentages.
Operating model
Which teams own the service once it runs?
Shift play-card production toward generate-then-review; keep teaching and tactical decisions with coaches; budget review time explicitly so ‘saved’ prep minutes are not double-counted.
Publication history
- 2026-09-15College Athletics · Issue 093 resources
Stable resource ID: asu-intel-football-play-card-agent-2026