High school student surrounded by police and handcuffed after AI mistakes his bag of Doritos for a gun

midian182

Posts: 11,628   +176
Staff member
WTF?! Even as billions are poured into AI and its use becomes increasingly widespread, we're still seeing very worrying instances of the technology getting things wrong. A prime example happened at a high school when a student was swarmed by police after an AI thought the Doritos bag he was holding was a gun.

Taki Allen, a 16-year-old high school student at Kenwood High School in Baltimore County, Maryland, got an unwelcome surprise as he sat outside with friends after football practice while waiting to be picked up last week.

About 20 minutes after he finished the bag of Doritos he had been eating, around eight police cars showed up. Officers climbed out of the cars and, with their weapons pointing at Allen, told him to put his hands on his head and walk toward them. The student said he was then told to get on his knees and was placed in handcuffs. Baltimore County Police Department later confirmed that Allen was handcuffed but not arrested.

Unfortunately for Allen, the artificial intelligence that reviews the school's surveillance camera footage mistook the crumpled Doritos bag for a firearm and alerted the police.

"Do you have a gun on you?" one officer asked the boy as the others searched his friends.

After reviewing the footage that the AI had identified as a gun, police looked in a nearby trash can and found the bag of chips.

"I guess just the way you were eating chips … Doritos, whatever …. it picked it up as a gun," one officer said. So, remember: never eat chips the wrong way. Another cop added that "AI is not the best," which is an understatement.

School Superintendent Dr. Myriam Rogers told reporters that the system worked how it was meant to. "The program is based on human verification and in this case the program did what it was supposed to do which was to signal an alert and for humans to take a look to find out if there was cause for concern in that moment," Rogers said.

Understandably, Allen didn't share Rogers' point of view. "I don't think no chip bag should be mistaken for a gun at all."

School principal Kate Smith sent a letter to parents following the incident emphasizing the importance of student safety, but Allen said Smith never spoke to him until three days after it happened.

The mistake has led to calls from local politicians for the school's use of AI-based surveillance technology to be reviewed.

Allen said he now waits inside after football practice, as he does not think it is "safe enough to go outside, especially eating a bag of chips or drinking something."

Omnilert, the developer of the AI system used by the school, told BBC News, "We regret this incident occurred and wish to convey our concern to the student and the wider community affected by the events that followed."

"While the object was later determined not to be a firearm, the process functioned as intended: to prioritise safety and awareness through rapid human verification," it said.

Permalink to story:

 
School Superintendent Dr. Myriam Rogers told reporters that the system worked how it was meant to.
Proof AI isn’t ready for real-world use, especially in schools. Any system that mistakes a chip bag for a gun and sends police on high alert isn’t “working as intended,” it’s a disaster.

This occurring once is too many times already. Only a pustule of a human being would claim otherwise.

Dr. Rogers’ statement is downright disgraceful. Defending any automated computer failure that put a kid’s life at risk as a success isn’t just wrong, it’s inhumane, and frankly, should be a career-ending statement.

There is only one decent reaction here: we will remove the system until it is provably improved so we don’t inadvertently terrorize another child in the future. Anything less is jaw-droppingly awful.
 
"After reviewing the footage that the AI had identified as a gun, police looked in a nearby trash can and found the bag of chips."

Why was this not done while the police were being dispatched? If the system can alert 911 on its own then it can provide a link to the video footage.
 
Proof AI isn’t ready for real-world use, especially in schools. Any system that mistakes a chip bag for a gun and sends police on high alert isn’t “working as intended,” it’s a disaster.

This occurring once is too many times already. Only a pustule of a human being would claim otherwise.

Dr. Rogers’ statement is downright disgraceful. Defending any automated computer failure that put a kid’s life at risk as a success isn’t just wrong, it’s inhumane, and frankly, should be a career-ending statement.

There is only one decent reaction here: we will remove the system until it is provably improved so we don’t inadvertently terrorize another child in the future. Anything less is jaw-droppingly awful.
How convenient you dont include the full quote.

"The program is based on human verification and in this case the program did what it was supposed to do which was to signal an alert and for humans to take a look to find out if there was cause for concern in that moment," Rogers said."

The AI did exactly what it was designed to do. It saw what it thought was a threat and created an alert. The obvious issue here is that none of the HUMANS involved, the security team in charge of the cameras, or the police, ece bothered to check the footage the AI alerted to before deciding to roll out the squad.

But it's easier to cry about AI being the root of all evil then to admit this is a policy failure at best, human oversight at worse. Insisting someone should have their career ruined over this is actually psychotic.
I feel sorry for the kids. Other first world countries do not have to suffer a gun fetish like America does.
Dont worry, they have their own fetishes to worry about.
"After reviewing the footage that the AI had identified as a gun, police looked in a nearby trash can and found the bag of chips."

Why was this not done while the police were being dispatched? If the system can alert 911 on its own then it can provide a link to the video footage.
That's what I'm saying. We have an AI detection system, and anything it detects is immediately forwarded to an actual human for review.

Did nobody look at this? Does it do it automatically? Or worse, did someone not even bother to check the footage and immediately hit the gas on escalation?
 
How convenient you dont include the full quote.

"The program is based on human verification and in this case the program did what it was supposed to do which was to signal an alert and for humans to take a look to find out if there was cause for concern in that moment," Rogers said."

The AI did exactly what it was designed to do. It saw what it thought was a threat and created an alert. The obvious issue here is that none of the HUMANS involved, the security team in charge of the cameras, or the police, ece bothered to check the footage the AI alerted to before deciding to roll out the squad.

But it's easier to cry about AI being the root of all evil then to admit this is a policy failure at best, human oversight at worse. Insisting someone should have their career ruined over this is actually psychotic.

Dont worry, they have their own fetishes to worry about.

That's what I'm saying. We have an AI detection system, and anything it detects is immediately forwarded to an actual human for review.

Did nobody look at this? Does it do it automatically? Or worse, did someone not even bother to check the footage and immediately hit the gas on escalation?

Well, firstly, the reviewer is likely another cop, someone who is trained to see a threat in everything. The AI did the equivalent of waiving a red flag in front of a bull, and we wonder why the bull decided to charge? Secondly, who wants to be the person that has to admit 'the AI flagged a threat and I discounted it, and now kids are dead'? With all the shootings and criticism police have had in the past decade for failing to stop school shooters quick enough, there is no way a cop would respond any other way then they way they did.

While the reviewer erred, the whole system has contributed to make such a mistake a near certainty. And that is the problem. So now we have the situation where having 8 armed police pull guns on an innocent teenager makes schools 'safer' because 'the system worked as intended'.

Yay, I am so glad we got here as a society, go us...
 
Only a matter of time and some one will be shot because of the Ai being so wrong it will cost some innocent person their life. Imagine if that boy had freaked out and ran becasue he was so scared, the police may have shot him.
 
Last edited:
"I don't think no chip bag should be mistaken for a gun at all."

Dr. Rogers must be delighted at the great job her school is doing.
 
"prioritise safety" by sending in police guns drawn with fabricated probable cause.

Another hidden benefit sending your kids to elite private school: police need a warrant and fear the powerful parents there. Can't afford it? Oh well then you aren't the safety being prioritized.
That's a really easy solution stop voting against school choice bills we need to put it into the public schools anyway in the absolute cancer of what the ideology they keep spouting
 
In a civilized country, not getting shot is just a given, not something you need to be thankful for happening. Here in Australia, anytime an police officer shoots their gun, there is an investigation to see if it was necessary. America is a sick country.
There are things wrong everywhere. I am glad you are in Australia (which America saved from Japanese conquest) since you think we are sick. Oh and here in the GREAT USA we investigate officer gun shootings also and better than Australia does. I know all about your country because I have watched Mr Inbetween several times and I love Judith Durham, may she rest in peace. Let me buy you a dimmie Mate.
 
Back