From the College Athletics edition of October 23, 2026
Texas Athletics launches Bevo Lab with Google Cloud, Dell and SHI: Gemini-based football scouting tool already credited with saving coaches 15+ hours a week; lab fully operational in 2027
University of Texas at Austin Athletics / McCombs Business of Sports Institute (with Google Cloud, Dell Technologies, SHI International; Learfield's Longhorn Sports Properties) · Athletic department performance analytics, roster strategy and student workforce development · United States (Austin, Texas; SEC)
- Publisher
- CBS Sports
- Original publication
- October 23, 2026, 9:09 a.m. ET
- Source retrieved
- 2026-10-24
- Event date
- 2026-10-23
What happened
On October 23, 2026, Texas Athletics and the McCombs Business of Sports Institute announced Bevo Lab, a 25,000-square-foot sports analytics and AI hub in Austin's Innovation Tower, with Google Cloud (official cloud and enterprise AI provider, Gemini), Dell Technologies (AI infrastructure) and SHI International (integrator) as founding partners. CBS Sports reports that football coaches already query a Gemini chatbot prototype about opponent tendencies, and that a tool called Football Scouting Intelligence automates opponent data and video reports staff used to compile by hand. Texas says early work with Google Public Sector has saved coaches more than 15 hours a week in game preparation; Kirk Goldsberry said the figure comes from the game-planning reports. A Transfer AI model rates roughly 5,000 men's and 5,000 women's college basketball players and informs revenue-sharing budget decisions. The release describes a unified 'Bevo Lakehouse' for game film and wearable data, starting with football and expanding to all 21 sports and 500+ athletes. Goldsberry has hired 24 undergraduates; up to 40 student researchers in year one, scaling to 200; construction about three months; fully operational in 2027.
Why it matters
Why now: announced October 23, 2026, the edition date, at Texas' Business of College Sports Summit. It is the largest athletics-wide AI build announced by a college department in the archive, pairing a hyperscaler, hardware and integrator sponsorship with a student workforce. It extends earlier archived Texas work with Equipe's conversational data tool (August 2026) into coaching, roster and revenue-sharing decisions. Interpretation: it gives other Power-conference departments a template, and a benchmark claim (15 hours a week) that buyers will hear repeated, so the claim's basis matters.
Evidence and measured results
CBS Sports reporting with on-record interviews (Del Conte, Goldsberry, Google Cloud's Matt Renner), corroborated by the university/Learfield release (PR Newswire copy and texaslonghorns.com) and a Google Cloud interview. The 15-hour figure is self-reported by Texas from its game-planning report process; no baseline method, staff count, accuracy check or outcome on results is published. Front Office Sports' interview records Goldsberry acknowledging sensitive student-athlete data and ethical concerns. A Yahoo Sports article the same day attributes a failed fourth-and-2 call against Florida to Bevo Lab, without saying how the recommendation was produced or delivered.
Limitations and uncertainty
Most capabilities are planned; the chatbot is a prototype and the lab is not fully operational until 2027. Time savings are self-reported. Sponsor-driven announcement with marketing language ('global benchmark'). NCAA football rules bar analytics on in-game tablets, so in-game use claims (e.g., the Yahoo report) need verification. Resourcing (hyperscaler engineers, 40 to 200 students) is not transferable to most departments.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-10-23; 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 staffs spend many hours each week compiling opponent and roster reports from disconnected film, data and wearable systems.
- Stakeholders
- Athletic director
- Head coaches and football/basketball analytics staff
- Deputy AD for sport performance
- University IT and data governance office
- Business school or analytics faculty
- Multimedia rights partner (sponsorship)
- Discovery questions
- Which weekly reports consume the most staff hours today, and how are hours measured?
- Where do film, wearable and scouting data live, and is there a common athlete ID?
- Who may access athlete performance and health-adjacent data?
- Would AI outputs inform revenue-sharing or roster decisions, and who reviews them?
- Is there academic capacity (students, faculty) to support a build?
- Value hypothesis
- Automating recurring scouting reports and unifying data could return staff hours to coaching, if outputs are accurate and the data foundation exists.
- Potential engagement
- A six-to-eight-week data-readiness assessment and single-report automation pilot for one sport, with a measured staff-hour baseline.
- Unsupported claims
- 15-hour weekly savings as a transferable benchmark
- Improved win rates or play-calling
- Accuracy of Transfer AI player ratings
- That a department can replicate Texas without hyperscaler sponsorship
Pre-sales engineering
Role takeaway
- Fit
- Large departments with multiple data sources, analytics staff and the ability to fund or sponsor a cloud data platform.
- Architecture and integration
- Cloud lakehouse consolidating game film, wearable and scouting feeds; LLM conversational layer (Gemini) over curated datasets; custom ML models; role-specific front ends built through APIs.
- Prerequisites
- Inventory of film, wearable and scouting systems
- Common athlete identifier
- Data-sharing and consent terms
- Access-control model for students and vendors
- Analyst-prepared reports to use as ground truth
- Constraints
- NCAA football rules prohibit analytics and data access on in-game tablets
- Health-adjacent data rules (HIPAA/FERPA where applicable)
- Vendor export and API limits
- Staff and student capacity outside large programs
- Security
- Role-based, credentialed access with audit logs; segregate medical and wellness data; vet student and partner-engineer access; define retention for athlete records.
- Proof of value
- Compare chatbot answers with analyst-prepared opponent reports on sample questions
- Measure staff hours on one weekly report before and after
- Validate transfer-model ratings against subsequent season performance
- Test that unauthorized roles cannot reach athlete data
Delivery
Role takeaway
- Work
- Phased consolidation of one sport's data, automation of a recurring report, chatbot pilot for coaches, then expansion by sport.
- Dependencies
- Coaching staff willingness to share processes
- Cloud and integrator partner capacity
- Student researcher pipeline and training
- Data-governance approval
- Ownership
- A named athletics analytics lead accountable to the athletic director, with university IT owning platform security.
- Skills and adoption
- Coaches learn to query data in natural language and to challenge outputs; students and analysts maintain data pipelines and models.
- Governance checkpoints
- Privacy and consent review before wearable or health-adjacent data is ingested
- Human review before AI outputs inform roster or pay decisions
- Confirm compliance with NCAA in-game technology rules
- Acceptance criteria
- Proposed: measured staff-hour reduction on a defined weekly report against a pre-pilot baseline
- Proposed: coach-validated accuracy on a sample of chatbot answers
- Proposed: documented access controls tested before expansion
- Risks
- Over-reliance on AI recommendations in high-stakes decisions
- Exposure of sensitive athlete data
- Sponsor-driven scope outpacing governance
- Student turnover disrupting maintenance
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?
Google Cloud data platform with Gemini as the conversational layer over a unified lakehouse of game film, wearable (e.g., Catapult) and scouting data; Dell-supplied infrastructure; SHI integrating the stack; student-built APIs and role-specific front ends for coaches, athletes and executives. Interpretation: the data consolidation, not the chatbot, is the hard prerequisite.
Governance
Who approves, reviews and stays accountable for outcomes?
Sponsorship arranged through Learfield's Longhorn Sports Properties; models feed revenue-sharing budget decisions; students handle athlete data. Goldsberry told Front Office Sports there is a blurry line between medical data and sports analytics and each university needs a serious ethics conversation. No published data-governance policy, consent terms or model-review process.
Security and privacy
What data, permissions and controls need testing?
Centralizes film, wearable and potentially health-adjacent data for 500+ student-athletes, accessed by student researchers and vendor engineers. Texas says access will be credential-restricted on Google Cloud; retention, athlete consent and athlete access rights are not described.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Creates paid roles for 24 undergraduates now and up to 40 a year (scaling to 200), working 10–20 hours a week per Goldsberry; shifts analyst time from report compilation to strategy.
Procurement
What should contracts, pricing and exit terms secure?
Delivered as corporate sponsorship arranged by Learfield's Longhorn Sports Properties rather than a disclosed purchase; contract values, terms and data rights are not disclosed.
Operating model
Which teams own the service once it runs?
Moves analytics from sport-by-sport silos to a central athletics-academic lab serving all 21 sports, staffed by students under a business-school institute.
Publication history
- 2026-10-23College Athletics · Issue 443 resources
Stable resource ID: texas-athletics-bevo-lab-google-cloud-dell-shi-ai-sports-intelligence-hub-2026-10-23