homehome Home chatchat Notifications


How AI analysis of millions of hours of body cam footage could reform the police

Most body cam footage stays unseen. But some scientists see an opportunity in leveraging this data.

Tibi Puiu
November 14, 2024 @ 10:00 pm

share Share

Illustration by Midjourney.

When police officers in Oakland, California, began wearing body cameras, the goal was clear: enhance transparency. Sparked by national outrage over the 2014 shooting of Michael Brown, the federal government, under President Obama, pledged $75 million to help law enforcement agencies adopt this technology. The idea was to restore public trust, especially in communities that felt targeted by police brutality. The cameras captured millions of hours of footage — traffic stops, arrests, even mundane conversations on street corners.

But the footage collected largely went unused. Beyond the courtroom and occasional media scrutiny, these clips sat in digital vaults, unseen by the public. As Benjamin Graham, a political scientist at the University of Southern California, notes for Scientific American, “We spend so much money collecting and storing this data, but it’s almost never used for anything.”

Researchers like Graham are trying to change that. By reimagining body camera footage not just as evidence but as a vast reservoir of data, they aim to transform how police departments operate. Using advanced artificial intelligence, these scientists are mining video transcripts to uncover patterns in officer behavior. This approach, they argue, could reshape police training and build trust in communities where that trust is frayed — and the results so far have been very promising.

Detecting Patterns in Policing: From Bad Language to Escalation

Jennifer Eberhardt, a psychologist at Stanford University, has spent years exploring police interactions. Together with Dan Jurafsky, a Stanford linguist, her team began to analyze body camera footage in 2014. The initial focus was on the Oakland Police Department, which had adopted the technology years earlier. For Eberhardt, the goal was to go beyond the surface level and understand, moment by moment, how police communicate with citizens.

Eberhardt’s team developed an AI-powered model to assess the respectfulness of language used by officers during traffic stops. They trained the system by analyzing nearly 1,000 transcripts, measuring how officers spoke to Black and white drivers.

The findings were eye-opening: officers spoke less respectfully to Black drivers than to white ones. They were less likely to explain the reason for the stop, offer reassurances, or express concern for the driver’s safety. These discrepancies existed regardless of the officer’s race, the reason for the stop, or the eventual outcome. When the Stanford team presented their results to the Oakland Police Department, it validated long-held suspicions among the city’s minority communities.

These insights didn’t just linger in academic journals — they prompted action. The Stanford researchers helped Oakland develop a “respect” module for their training programs, drawing directly from real-life interactions recorded by the cameras. Officers were trained to use more respectful language and explain their actions to those they interacted with. The effects were tangible: after implementing the training, officers became more likely to reassure drivers and provide clear reasons for stops.

But the researchers didn’t stop there. In a separate study last year, they analyzed body camera footage to identify “linguistic signatures” that predicted when a traffic stop would escalate into an arrest or search. By scrutinizing the first 45 words spoken by officers, they noticed a pattern. Apparently, giving direct orders without explanation often led to an unfavorable outcome for the driver. This pattern was disturbingly present in the footage from George Floyd’s fatal encounter with police officers in 2020.

A Data-Driven Path to Reform or a Political Minefield?

Inspired by these findings, police departments in Los Angeles and San Francisco are now exploring similar projects. The Los Angeles Board of Police Commissioners, for instance, enlisted Graham’s team at U.S.C. to analyze 30,000 body camera videos, aiming to understand the nuances of traffic stops. The San Francisco Police Department has also teamed up with the Stanford researchers to assess a program focused on nonviolent communication.

However, expanding this approach faces significant hurdles. Departments may hesitate to open their footage to scrutiny, fearing what these analyses might reveal. In some cases, the reluctance stems from concerns over privacy or the potential for legal repercussions. But as Eberhardt notes, systematic analysis of these interactions is crucial to understanding the real impact of police training. “A lot of those trainings that they have now are just not evaluated rigorously,” she told SciAmerican. “We don’t know whether whatever it is that they’re learning… actually translates to real interactions with real people on the street.”

“By taking on these types of studies and making improvements in your department, it helps actually to build trust in communities that have really low trust levels,” added LeRonne Armstrong, former chief of police of California’s Oakland Police Department.

share Share

This Film Shaped Like Shark Skin Makes Planes More Aerodynamic and Saves Billions in Fuel

Mimicking shark skin may help aviation shed fuel—and carbon

China Just Made the World's Fastest Transistor and It Is Not Made of Silicon

The new transistor runs 40% faster and uses less power.

Ice Age Humans in Ukraine Were Masterful Fire Benders, New Study Shows

Ice Age humans mastered fire with astonishing precision.

The "Bone Collector" Caterpillar Disguises Itself With the Bodies of Its Victims and Lives in Spider Webs

This insect doesn't play with its food. It just wears it.

University of Zurich Researchers Secretly Deployed AI Bots on Reddit in Unauthorized Study

The revelation has sparked outrage across the internet.

Giant Brain Study Took Seven Years to Test the Two Biggest Theories of Consciousness. Here's What Scientists Found

Both came up short but the search for human consciousness continues.

The Cybertruck is all tricks and no truck, a musky Tesla fail

Tesla’s baking sheet on wheels rides fast in the recall lane toward a dead end where dysfunctional men gather.

British archaeologists find ancient coin horde "wrapped like a pasty"

Archaeologists discover 11th-century coin hoard, shedding light on a turbulent era.

Astronauts May Soon Eat Fresh Fish Farmed on the Moon

Scientists hope Lunar Hatch will make fresh fish part of space missions' menus.

Scientists Detect the Most Energetic Neutrino Ever Seen and They Have No Idea Where It Came From

A strange particle traveled across the universe and slammed into the deep sea.