About a third of Major League Baseball teams had quietly built something new into their dugout iPads: custom apps that watched every pitch in real time and told coaches what to throw next. The Miami Marlins got there first, and the practice spread to as many as six more teams. The New York Mets, under owner Steve Cohen, reportedly spent several hundred thousand dollars on a system that helped pick pitches from the dugout. On July 16, MLB shut it down. Executive VP Morgan Sword blocked the third tab on league-issued tablets — where teams had installed their AI tools. He told all 30 clubs the technology had gone "beyond its originally intended purpose." MLB punished no one — the teams had broken no existing rules. But the debate over what happened, and what it means for the game, is just getting started.
The Game's Human Element Is Non-Negotiable
Real-time AI in the dugout isn't just an analytics tool — it's replacing the judgment that defines baseball.
Pitch-calling is the beating heart of the game. The back-and-forth between pitcher and catcher, manager and bench, is what separates baseball from most sports. When an algorithm is making those calls in real time, the humans holding the signs just execute — they don't decide. That's what bothered Sword: the iPads had moved beyond replay and stats into live recommendations on substitutions, pitch calls, and in-game decisions that players and coaches are supposed to make.
Players and managers both think AI crossed the line. Yankees captain Aaron Judge said he "can't believe" teams are making decisions off AI: "Man, that's just crazy." Blue Jays manager John Schneider named pitch-calling as the specific worry — the way AI could "sway your decision one way or another" in real time, in a way that undermines what a manager actually does. Jack Lambert of Driveline Baseball, one of the sport's leading analytics firms, supports the restriction for the same reason: the gap between data-informed preparation and live AI decision-making isn't cosmetic, and the league was right to draw it.
The competitive equity concern is real too. Steve Cohen's Mets reportedly spent several hundred thousand dollars on their system. That's an easy check for a billionaire owner — not so easy for smaller-market clubs. An unnamed league executive put the logic of the ban bluntly: "Gotta stop the cheating before there's cheating now." Even without a formal rules violation, the direction was clear.
But ... AI Is Just the Next Wave of Analytics
Blocking AI pitch-calling treats this generation's analytics tool as uniquely illegitimate — and that argument doesn't hold.
Statcast was controversial once too. When MLB introduced ball-tracking and exit-velocity data in 2015, traditionalists worried it would reduce the game to numbers. It didn't — it gave teams better information, and the best managers still had to interpret and act on it. Arizona Diamondbacks manager Torey Lovullo said his team wasn't using it — but understood exactly why others did: "It hasn't impacted us at all but I know AI is entering our arena for sure. It's entering everyone's arena. You better get on it, or you're going to get rolled over by it." Banning it doesn't protect the game. It just delays the inevitable while other sports get ahead.
The teams broke no rules. MLB confirmed that after its internal review. They weren't cheating — they were being creative within the rules that existed. A senior front-office executive at one of the affected clubs told The Athletic it "caused quite the stir." The Marlins built it. The Mets invested in it. About ten clubs adopted some version of it. That's not a fringe behavior — it's the league's analytical vanguard moving in the same direction.
The ban also draws an odd line. Teams can still use pre-game AI models, Statcast dashboards, and historical tendency data on those same iPads. The league's line: preparation is fine, live recommendations are not. But in practice, the richest clubs will just move their AI work into pre-game packages — and policing the real-time edge gets harder, while disguising it gets easier.
Where This Lands
The camps aren't really disagreeing about technology. They're disagreeing about what baseball is. If the game is ultimately a test of human judgment under pressure, live AI recommendations break the game. If it's a competition between organizations — from scouting to front offices to analytics departments — then banning AI is like telling one era's Bill James disciples they went too far. MLB drew the line at the in-game moment. Whether that line holds, or just moves somewhere harder to see, is the real question.
Sources: [Fox Sports/AP](https://www.foxsports.com/articles/mlb/mlb-restricts-dugout-ipad-use-to-prevent-ai-help-with-strategy-ottavino-says-mets-were-involved) · [Chicago Sun-Times](https://chicago.suntimes.com/mlb/2026/07/17/mlb-restricts-dugout-ipads-to-bar-use-of-ai-to-make-game-decisions) · [NBC Sports](https://www.nbcsports.com/fantasy/baseball/player-news/2026-07-16/mlb-outlaws-generative-ai-in-dugouts) · [Yahoo Sports/The Athletic](https://sports.yahoo.com/articles/gotta-stop-cheating-cheating-mlb-113022136.html) · [Yardbarker](https://www.yardbarker.com/mlb/articles/mlb_cracks_down_on_dugout_ai_loophole/s1_17701_44074117) · [Bleacher Report](https://bleacherreport.com/articles/25455579-mlb-reportedly-outlaws-use-dugout-ipads-access-ai-game-strategy) · [Albat](https://en.albat.com/thebigs/mlb-mets-break-their-silence-with-bold-response-to-artificial-intelligence-ai-controversy-20260718-0066.html) · [Operation Sports](https://www.operationsports.com/former-met-adam-ottavino-says-the-mets-were-using-an-expensive-ai-program-to-predict-pitches-from-the-dugout-before-mlb-banned-the-technology/) · [Engadget](https://www.engadget.com/2217032/mlb-bans-using-dugout-ipads-for-ai-powered-in-game-strategy-calls/) · [IBTimes AU](https://www.ibtimes.com.au/mlb-bans-ai-dugout-ipads-1872469)