MLB partners with Uplift Labs, which uses AI and images captured by two iPhone cameras to detect players' flaws, forecast their potential, and flag injury risk
This is the real AI threat, hidden forms of bias become encoded into systems in ways that are hard to observe and fix. — https://www.wsj.com/... Twitter: @upliftlabs : We are thrilled to announce Uplift Labs' AI technology will be deployed across Major League Baseball for large-scale prospect evaluations including the MLB Draft Combine, as highlighted in the Wall Street Journal. #sportstech https://www.wsj.com/...
Context & Ripple Effects
MLB's move from experimental to official AI is the story here. As early as the 2020 wave of machine learning in sports, teams were using models to assess talent and flag injuries, and by 2022 computer vision was already predicting injuries and shaping tailored workouts at the coaching level. What changes now is scale and sanction: the league itself is putting Uplift Labs' two-iPhone system at the MLB Draft Combine, making an algorithm a formal input into who gets drafted.
It also lands inside a league that is simultaneously normalizing cameras elsewhere — the Sony-camera Automated Ball-Strike System backing up human umpires — while walling off some uses, notably the ban on league-provided dugout iPads tapping GenAI for in-game strategy. MLB is drawing a line between AI as sanctioned infrastructure and AI as unauthorized advantage.
First-order effects
- Draft Combine prospects are now screened through a standardized AI pipeline — mechanics, projected potential, and injury risk assessed from two consumer iPhone cameras rather than bespoke lab equipment.
- Uplift Labs converts an NFL-and-NBA-style niche tool into league-wide distribution across all 30 clubs at once, the single largest deployment in its market.
Second-order effects
- Scouting departments face pressure to justify human evaluations against model outputs, shifting draft-room leverage toward analysts who can interrogate the system — and toward clubs that can contest its flags most credibly.
- Competing leagues already experimenting with AI tactics, like Liverpool's work with DeepMind, gain a template for league-sanctioned deployments, while sensor-hardware vendors lose ground to commodity smartphone capture.
Third-order effects
- The WSJ's own framing — hidden forms of bias encoded into systems that are hard to observe and fix — points to a structural problem: when a league-standard model shapes who enters professional baseball, flawed training data becomes career-altering for teenagers, inviting future audits of algorithmic talent evaluation much as officiating tech faced scrutiny.
- If the pattern holds, leagues consolidate around a few officially licensed AI vendors, turning player evaluation from a club-by-club edge into shared infrastructure whose errors are systemic rather than local.
The trend: Professional sports leagues are converting computer-vision tools from optional coaching aids into official, league-licensed infrastructure — deciding where AI is sanctioned and where it is banned.