Higher Education & ResearchIssue 482 records
College Athletics
Public records reported by FOIAball on Oct 30 show football programs' direct AI subscriptions are small, from $20 a month for Claude at Kansas and Virginia Tech to about $3,700 a year with Anthropic at Michigan State, and that line items miss AI built into sports-tech vendors and campus licenses. A Florida Undergraduate Law Review post (Oct 27) argues athletes have no reliable way to see or contest algorithmic assessments drawn from wearable data. Interpretation: athletics AI use is growing faster than departments can inventory it or explain it to athletes.
- Evidence records
- 2
- Cross-source patterns
- 0
- Evidence classes
- 1 independent reporting1 independent research
- Outcomes
- 1 emerging1 cautionary
- Source freshness
- 2 new this fortnight
- Research completed
- 2026-10-31
Choose a role to see its takeaway beside every record in the ledger.
Patterns across the evidence
Synthesis · Lighthouse Advisory interpretation
The evidence in this edition did not support a cross-source pattern. Each record below stands on its own.
Evidence ledger
Full record · every source keeps its link and limitations
Public records show football programs' direct ChatGPT and Claude spending is small and opaque (Kansas, Virginia Tech, Boise State, Michigan State, Texas Tech, Iowa)
FOIAball requested every university athletics department's spending with Anthropic and OpenAI after seeing a ChatGPT charge in an Iowa expense sheet. Records from a few schools show small direct football subscriptions: Kansas $20 a month for Claude Pro, Virginia Tech $20 a month for Claude, Boise State $75 a month, Michigan State $3,700 a year with Anthropic, and Texas Tech about $3,500 a year with Anthropic plus $300 a year for ChatGPT. Iowa spent $312 on ChatGPT, period unclear; Iowa had not yet responded. The author notes these payments miss AI built into vendors such as Catapult and Hudl and AI access bundled into campus-wide Microsoft, Google, OpenAI or Anthropic deals, and that the records say nothing about how the tools are used.
Why it matters, evidence and limitations
- Why it matters
- Why now: published Oct 30, a week after Texas launched Bevo Lab without a price tag. It is the first records-based look at what football programs pay AI model providers directly, and it shows that line-item spending is a poor measure of athletics AI use: most AI arrives through sports-tech vendors or campus licenses.
- Evidence and measured results
- Public-records responses described by one reporter; documents are not published or linked; sample is 'a few schools' with no response rate given. Figures are subscription amounts only. The comparisons with team records are satire, not analysis.
- Limitations and uncertainty
- Single satirical newsletter; small, self-selected sample; amounts only; no usage information; underlying records not public; Iowa period unclear.
Florida Undergraduate Law Review: athletes lack a way to see or contest algorithmic assessments that shape clearance, playing time and scholarships
The post argues that athlete performance and biometric data fall across FERPA, the Fourth Amendment at public universities, uneven state biometric laws and non-binding NCAA guidance, with no consistent rule. When software estimates injury risk from wearable data, an athlete can ask to correct education records under FERPA but cannot overrule a coach and has no guaranteed access to the algorithm's logic. It notes consent to wearing monitors may not be voluntary when refusal could cost playing time. It proposes making the NCAA's March 2026 performance-technology recommendations binding, with annual compliance documentation, athlete access to their data, a process to contest algorithmic assessments, vendor contract protections, and no sole reliance on automated assessments for medical clearance or scholarships.
Why it matters, evidence and limitations
- Why it matters
- Why now: published Oct 27 as departments such as Texas build AI models on wearable and film data for roster and transfer decisions. It frames a concrete control gap that departments adopting predictive athlete models will need to address: notice, access, appeal and human review.
- Evidence and measured results
- Legal commentary drawing on FERPA regulations, case law, NCAA's 2026 recommendations and prior law-review articles; no empirical data or documented cases.
- Limitations and uncertainty
- Undergraduate blog commentary; not peer-reviewed; no empirical evidence that schools make decisions this way; recommendations are the author's.
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.
- Independent research
- Research conducted outside the implementing organization.