CrimeRadar has apologized for the distress caused by false crime alerts issued to local US communities after a BBC Verify investigation.

[dimab]

AI app causes distress across US with false crime alerts

An AI-powered app called CrimeRadar has apologised for the distress caused by false crime alerts issued to local US communities after a BBC Verify investigation.

www.bbc.com

 

[Enforcer52]

I have been monitoring the alerts sent by CrimeRadar that come from my feeds. Although none have caused the panic like the Pilot Point alert last week, CrimeRadar's AI continues to butcher the translations of the actual calls, I guess it doesn't understand the Texas Drawl accent.

I fear it is only a matter of time before before a misleading headline on one of my feeds causes an area wide panic.

The useless mundane EMS alerts over leg pain, diarrhea, vomiting, nose bleeds, etc. is just ridiculous!

Just to continue saying they are "sorry" and that they are "working" on the software, doesn't cut it for me.

 

[blantonl]

What is happening here is not constructive AT ALL. Crimeradar is not being given a fair opportunity to address the concerns being raised, instead, there has been a rush to publicly trash the platform rather than engage in good-faith discussion by a small group of people here. I can state firsthand that Crimeradar is highly committed to improving their product quality and has already taken immediate action to remove sensitive talkgroups. These concerns have been escalated to, and are being actively addressed by their senior leadership and engineering.

In addition, Crimeradar has currently been having productive, direct conversations with multiple feed providers and with agencies themselves, discussions that were professional and focused on solutions.

Unfortunately, this pattern is not new around here. Every time there is a change or new technology in this hobby, whether encryption, AI, interface design, or otherwise, a vocal subset reacts like children and takes to public forums to complain rather than contribute constructively. It is counterproductive and, frankly, exhausting. The train is moving forward, with or without you.

If the intent is simply to complain and attack other organizations, this is not the place for it. Those who are willing to engage professionally, adapt to change, and provide thoughtful, constructive feedback are welcome and encouraged to do so. Otherwise, take your ball, feed, casserole, whatever, and move on.

 

[KD0FEO]

I can state firsthand that Crimeradar is highly committed to improving their product quality and has already taken immediate action to remove sensitive talkgroups. These concerns have been escalated to, and are being actively addressed by their senior leadership and engineering.

It's nice to have this update and knowing that it isn't being pushed under the rug. Are you able to provide an update as to what falls under the sensitive talkgroups? Law-Tac, Fire-Tac? The Fire-Tac raised hell in my county because they do mock scenarios and rescues for training and AI thought it was real, received messages on Facebook for almost 2 hours after the notification was broadcast by the app on our scanner page. That is why I appear frustrated, apologies for being a pain in the ass about it.

 

[IcomIcR20]

It's nice to have this update and knowing that it isn't being pushed under the rug. Are you able to provide an update as to what falls under the sensitive talkgroups? Law-Tac, Fire-Tac? The Fire-Tac raised hell in my county because they do mock scenarios and rescues for training and AI thought it was real, received messages on Facebook for almost 2 hours after the notification was broadcast by the app on our scanner page.

I was told that Law Tac and Law Talk service types were blocked across the board, and I have not seen any alerts for such talkgroups since that change was implemented. I provided some specific suggestions to Crime Radar on how to minimize issues with training exercises and was told that my feedback would be passed on to the engineering team. While I do have a number of concerns with regard to the Crime Radar platform, to their credit, they have seemed to be quite receptive to my feedback.

 

[blantonl]

I'll say it again. We have asked SEVERAL TIMES for options to opt out because they have not done anything to mitigate the issue. We provide the scanner feeds to you, we should have at least some sort of say in how they are handled. Without feed providers, you wouldn't have Broadcastify.

I censored the post above, but I'm going to respond publicly to this comment this poster made by reposting what I said in an earlier thread with some extra emphasis.

The relationship between feed providers, premium subscribers, and the platform is and has always been mutually beneficial.

Participation is entirely voluntary ​

No one here is compelled to provide a feed, and no one is obligated to continue to be a premium subscriber. If you believe that relationship is no longer mutually beneficial, you are free to move on. Where issues most often arise is when a feed provider or subscriber all the sudden believes they are entitled to disproportionate influence or assumes the platform is acting in bad faith or for outsized financial gain.

I welcome professional and constructive feedback, and I review and listen to all feedback, even the unprofessional. But the minute you start ordering me around and telling me how I need to run this platform is the minute I discount your opinions and I move on.