image copyrightNexpher Images via Getty Images
"Grand Theft Auto" (GTA) depicts a dirty and disordered dystopian world, which does not seem like a place where people would particularly care about traffic rules.
The popular game series has always featured players racing around vast cities in stolen vehicles. Evading police pursuits at high speeds, drifting, skidding and crashing are all part of the thrill of the game. Players can even ram directly into pedestrians on the road if they wish.
So it might come as a surprise to learn that the game series known for its vehicular mayhem is now being used to improve real-world road safety.
From testing bicycle safety technology to training autonomous vehicles to recognize and avoid road hazards, researchers have been turning "GTA" into an experimental platform through various modifications and transformations, hoping to reduce traffic accidents and road injuries.
This article contains content provided by Google YouTube. Because these contents use technology such as cookies or cookies, we will seek your consent before loading any content. You may wish to read the Google YouTube Cookie Policy and Privacy Policy before giving your permission. If you wish to read the above, please click "Accept and Continue".
Ending YouTube post
End of content
“We wanted to find a way to test wearable devices and other new digital traffic safety devices in situations that would be impossible to reproduce on real roads and would be too dangerous,” says Milena Jäger, a researcher in engineering and computer science at Osnabrück University of Applied Sciences in Germany.
The latest generation of GTA games uses a realistic physics simulation system, coupled with rich road environments, weather changes and traffic conditions, providing researchers such as Yeager with a low-cost and ready-made test platform for early-stage research. Even advanced driving simulators have difficulty replicating such environments, she points out.
Pavlo Bazilinskyy, a human factors engineer at Eindhoven University of Technology in the Netherlands, added:
"A mature product with a development budget of 200 million to 1 billion euros (223 million to 1.113 billion US dollars; 1.485 billion to 7.423 billion yuan; 7.335 billion to 36.675 billion NT dollars), its simulation effect far exceeds that of general research simulators."
End of Hot Reading
Bazilinski has used "Grand Theft Auto V (GTA V)" to study the eye movement patterns of road users in different dangerous traffic situations, hoping to provide a reference for future road design decisions.
A cyclist’s “sixth sense”
When Yeager and his colleagues developed a device to help cyclists and pedestrians detect dangers around them on busy roads, they chose to use the "GTA" series of games as a test platform.
One of the devices can be mounted on a bicycle handlebar. When a vehicle approaches from behind, the device will send a warning to the bicycle user through vibrations, and the intensity of the vibrations will gradually increase as the vehicle gets closer.
Another version is a vest that produces vibrating reminders - a "vibrotactile". Researchers say this device allows users to sense approaching vehicles as if they have a "sixth sense."
However, testing this type of technology on real roads is difficult because researchers don't want to put participants at risk.
Therefore, the research team thought of using "mods" of "GTA V", which are small software that modify the game's functions, allowing them to control the player's starting position and specify different test situations.
Prof Julius Schöning, professor of engineering and computer science at Osnabrück University of Applied Sciences in Germany, who led the research, said: “Since ‘GTA V’ does not ask players to obey traffic rules, almost any traffic situation can be deliberately recreated.”
The research team set up a bicycle in the laboratory, with a screen showing the "GTA V" screen in front and another screen behind it to simulate the rear view when the rider looks back.
Afterwards, the researchers arranged for the participants to "bike" on a section of road in the game's fictional city of Los Santos, while a large number of trucks would pass by at high speed.
The results show that both devices can effectively alert bicycle users to approaching dangerous vehicles. However, these technologies have yet to be developed into commercial products.
image copyrightJan Kettler
Test human behavior
However, in many traffic accidents, the root of the problem is not technology per se, but human behavior.
Therefore, the key to understanding how to improve road safety is to understand why people react in certain ways to potentially dangerous situations.
Bazilinski and his team wanted to find ways to track cyclists' reactions to hazards in busy urban environments.
They combined eye-tracking technology with "GTA V" to accurately record the movement of participants' eyes while riding bicycles in the game, and how they observed various dangerous scenes, such as a car leaving a parking space, a car door suddenly opening, or approaching a busy intersection.
"The goal of this research is to measure human behavior to assist future mobility systems and policy development," Bazilinski said.
He pointed to road design as one example.
"Assisting road layout decisions is a good example of application, with the ultimate goal of making roads safer and reducing traffic deaths and injuries."
Training self-driving vehicles
However, the most widespread use of the GTA series in road safety research is in training artificial intelligence (AI) to learn to drive safer.
At the Korea Advanced Institute of Science and Technology (KAIST) in Daejeon, South Korea (South Korea), engineers used "GTA V" to build a huge database covering a large number of in-game traffic accidents and other road situations.
The research team then used this data to train artificial intelligence (AI) algorithms, allowing the system to learn to identify dangerous vehicles on the road, thereby helping self-driving cars learn to avoid potential risks.
Many research teams around the world have also been using the "GTA" series of games to train artificial intelligence software to drive safely on busy city streets and avoid hitting various obstacles.
However, the biggest challenge for self-driving cars may not be simply avoiding collisions, but how to make judgments when faced with difficult choices.
For example, when a motorcycle carrying two people is approaching, should a self-driving car drive onto the sidewalk to avoid the motorcycle, but risk knocking over an elderly woman who is walking? Or should it crash into the motorcycle, putting the passengers in both vehicles at risk of injury?
Researchers at Eastern Michigan University in the United States are trying to use "GTA V" to deal with such problems and test an ethical decision-making algorithm for self-driving cars.
The research team designed nine different emergency situations in GTA V for testing.
The results showed that the algorithm could reduce the number of potential deaths by about 24% by weighing different moral dilemmas. However, this type of technology remains a controversial area of research.
As the launch date of the latest work in the "GTA" series, "GTA VI", approaches on November 19, 2026, researchers are already looking forward to how to use its simulated world to conduct new experiments.
However, to do this, they still need to obtain authorization from the game's developer, Rockstar Games.
"Its potential as a research tool is still huge," says Romain Delgrange, who has used the game to study human navigation at the Laboratory of Applied Psychology and Ergonomics in Paris.
(Rockstar Games and its parent company Take-Two Interactive did not respond to requests for comment from the BBC.)
This article was originally written in English. We used artificial intelligence to assist with translation and was reviewed by BBC journalists before publication. Learn more about how we use artificial intelligence.
Explore more
More short films
More short films end




