# Fort Worth Police Train With AI De-Escalation Simulator

> Fort Worth officers are testing an AI system that simulates tense encounters and scores their communication in real time.
- Title: Fort Worth Police Train With AI De-Escalation Simulator
- Summary: Fort Worth officers are testing an AI system that simulates tense encounters and scores their communication in real time. AI de-escalation training lets…
- Keywords: fort worth police department, artificial intelligence, de-escalation training, university of texas at arlington, shuchi deb, neurodivergence, police training, interactive simulation, officer communication, department of justice funding, news, politics
- Source: Dallas Express — https://dallasexpress.com/tarrant/fort-worth-police-turn-to-ai-for-real-time-de-escalation-training
- Author: Dallas Express
- Published: 2026-09-22T16:00:14+00:00
- Read time: 2 min
- Topics: fort worth police department, artificial intelligence, de-escalation training, university of texas at arlington, shuchi deb, neurodivergence, police training, interactive simulation, officer communication, department of justice funding, news, politics
## Fort Worth tests AI training tool
Fort Worth police are testing an artificial intelligence system developed at the University of Texas at Arlington to practice de-escalating tense encounters. The AI simulates interactions with people who may be autistic, neurodivergent or have disabilities.
## AI adjusts behavior based on officer choices
The system creates different difficulty levels, from compliant subjects to people impossible to de-escalate. The AI changes its behavior depending on how officers handle each interaction.

> We can create lots of different difficulty levels for the officers who are training. Some of the AI can be very compliant, some of them are not that compliant, some of them can be impossible to de-escalate.
## Real-time scoring of officer statements
The AI evaluates what officers say and scores individual statements as positive or negative. Officers see assessments through headsets in real time and get detailed analysis after the exercise.
## System recognizes connection and empathy
The technology identifies when an officer is listening, being kind or finding common ground with the simulated person. It also detects when an officer fails to establish connection and adjusts the scenario accordingly.
## No human role players needed
Officer Jorge Lopez, who helped develop the training, says AI eliminates the need for physical role players and removes safety concerns. This allows officers to train more frequently.

> With AI, it takes out the need for a physical human role player, it takes out safety considerations because there's no physicality to it.
## Preparing for hidden disabilities
Officers often don't know beforehand if someone has autism or another disability requiring a different communication approach. The simulations let officers recognize behavioral cues and practice adjusting responses safely.
## Different from surveillance AI
This AI training tool differs from police uses of artificial intelligence for surveillance, facial recognition or suspect identification. Here, AI functions as an interactive training partner for high-pressure encounters.
## Department of Justice funds the project
The Fort Worth police work is sponsored by the department and funded through the U.S. Department of Justice. UTA Associate Professor Shuchi Deb leads the research with Officer Nathan Owens as project manager.
## Part of broader police modernization
Police Chief Eddie Garcia has emphasized technology expansion and officer training during his tenure. This AI project complements other recent initiatives like First Amendment refresher training.
## AI as virtual training partner
Instead of analyzing evidence after incidents, this AI creates encounters before they happen. Officers can practice with cooperative subjects, difficult people and impossible scenarios, then review how the system scored their responses.
## Key takeaway

AI de-escalation training lets police practice high-pressure interactions repeatedly without human role players, receiving instant feedback on their communication choices.