The “AI Homeless Man Prank” — in which TikTok users use AI-altered photos to mimic a home invasion — has gone viral on social media, prompting laughs among content creators and their followers.

However, police departments across the country and in the U.K. are issuing warnings about the “stupid and potentially dangerous” trend. 

In one TikTok video with 65,000 views, a frantic mother pleaded via text for her daughter to call the cops after the young woman shared images of a gaunt-looking man standing in their kitchen and sleeping on the couch. 

In another video with more than 880,000 views, a terrified father called his son 21 times after he shared videos of a man with a straggly beard using the dad’s toothbrush and sleeping on his bed. “JOE PICK UP THE PHONE,” the dad wrote via text, according to the TikTok video. “DO NOT USE MY TOOTHBRUSH.”

They’re just two of the more than 1,500 videos that pop up under the tag “#homelessmanprank” on TikTok. While the content is shared by young users, adults are also participating. In a recent Good Morning America episode, host Michael Strahan admitted that his assistant even sent him a manipulated image of a strange man in his office. 

“I freaked out,” Strahan said on air, joking that his assistant is now looking for a new job. “You don’t know. It looks so real.” 

On Monday, Oct. 6, the City of Salem Police Department issued a Public Service Announcement about the trend. After receiving the image, many recipients “sincerely believed that there was an actual intruder in their home and called 911 to report a burglary or breaking and entering in progress, which necessitated an immediate police response,” the Salem Police Department said.

The department said that beyond “being in bad taste,” the pranks could potentially cause harm.

“This prank dehumanizes the homeless, causes the distressed recipient to panic and wastes police resources,” authorities wrote. “Police officers who are called upon to respond do not know this is a prank and treat the call as an actual burglary in progress thus creating a potentially dangerous situation.”

Similar warnings have been issued by police departments throughout the globe, according to the BBC. 

The Yonkers Police Department urged parents and guardians to teach children about responsible AI use. Officials also advised people to “pause before calling 911.” 

“Ask questions. Make sure it’s real,” the department wrote in an advisory on Facebook on Friday, Oct. 10. “Make sure your family members know you’re going to call 911, and to tell you THEN if it’s a prank, before it’s too late.”

While AI technology that creates videos and images normally has a watermark indicating its origins, the mark can easily be taken off by users, according to NBC News. It’s also difficult to identify where the AI-generated images come from. 

Though some teens have enjoyed the attention that comes from their videos, others are facing jail time in states where sharing false information with authorities is a criminal offense.

Never miss a story — sign up for PEOPLE’s free daily newsletter to stay up-to-date on the best of what PEOPLE has to offer, from celebrity news to compelling human interest stories.

Earlier this month, the Brown County Ohio Sheriff’s Office announced that two separate incidents took place within in 24 hours in which young people used AI to “create and report false home invasions.” Following the incidents, both juveniles were “criminally charged” for their part in the incidents. 

The department said, “We want to be clear: this behavior is not a ‘prank’ — it is a crime.”