The injured teenage survivor of a January 2025 shooting at a Nashville, Tennessee high school recently sued the manufacturer of an “AI gun detection” system that failed to detect the handgun that left two dead, including the shooter.
According to the lawsuit, which was filed in Davidson County court last month, the security company Omnilert either knew or should have known that there were “significant operational limitations in its gun detection system that could result in detection failures during actual emergencies, including limitations based on camera placement, proximity of the weapon to camera sensors, camera angle, lighting, and weapon visibility.”
Omnilert cofounder Ara Bagdasarian declined Ars’ invitation to answer questions about the lawsuit. System Integrations, the other defendant in the case, which resold the Omnilert system, also did not respond to Ars’ request for comment.
AI can’t even detect a fully formed gun, and California thinks the AI will have no problems detecting gun parts from just gcode??
Soooooo… which is going to ultimately turn out to be more effective way of detecting guns: AI systems, or dowsing? Both are gonna suck, be wholly inadequate for the purpose, and be giant wastes of public money - but obviously, but I’ve gonna admit, AI at least as mild possibility of being better than random chance. In optimal conditions. Maybe.
(You may be thinking “pfffft, surely people wouldn’t be stupid enough use dowsing rods to detect weapons, that’s just so clearly stupid”, but they did, and this “solution” was sold to them by slick conmen. …Sounds familiarrrr???)
So once again the United States has attempted a complicated technical solution to a legal problem.
Why don’t you just implement safe gun laws. You don’t even have to ban people from owning guns, although that would be a good idea. You just need to have basic background checks on gun purchases.
But then how will we go on our killing sprees? Murder our loved ones? Become cops? Do you just hate american culture?
Background checks are good, but they aren’t a solution to school shootings. Those are almost all parents giving kids guns or having shitty storage practices.
So charge the parents then
That doesn’t prevent anything and doesn’t change the system that produces these results in the slightest.
Perhaps a tendency not to give your kids guns would be part of the background check, other countries managing.
Oh 100% American gun culture is absolutely insane
American gun culture also is too broad a term to mean anything. Gun culture in my north eastern state is worlds apart from gun culture in Florida, which is worlds apart from gun culture in Massachusetts. It’s a big country and our culture is not homogenous across all 50 states and neither are our gun laws.
Why is this any better than a metal detector?
Asking the real questions here. My guess would be: they didn’t have metal detectors, the metal detectors they had reached end-of-life, or preexisting metal detectors failed to integrate into a modern, unified surveillance system. And so the use of AI analytics tools, atop (preexisting) camera systems seemed more hassle-free (a subscription-based software integration) and cost-effective in the short term; that is if the unproven compromise bares any trust…
Metal detectors in schools are dystopian and nobody who works in a school wants them.
AI is so much better and not dystopian 🙏
The detection processing is completely invisible. It’s just a tool. How it’s used is what determines if it’s dystopian or not. Seeing the state the US is in, I’d argue the country itself is dystopian and this system is trying to somewhat protect people from that dystopia.
Soon: ai not meant to detect guns or prevent shootings, read fine print Court: OK cool, case dismissed
I guarantee their contract already has a legal disclaimer for this exact scenario.
I am so tired of AI being shoved into everything and then people surprised when it doesn’t work. There’s no AI I think that could have detected a small firearm easily concealed. Hell as it is with legal concealed carry you can’t tell who is legally carrying as it is even with some of the most observant eyes watching.
There’s no AI I think that could have detected a small firearm easily concealed.
The idea with these kinds of systems are meant to allow early warning when possible.
No system is going to be 100%.
Edit: I get the downvotes, but there are people/companies that were/are developing such systems with a genuine intent to make things better. I know, because I was one of them.
AI in this context is useless though, you could paint marker “not a gun” on the side of a gun and guess what would happen.
It has some uses, but 95% of what is being used for and 100% of the data centers aren’t it.
you could paint marker “not a gun” on the side of a gun and guess what would happen.
It would flag it as a gun. How do I know? I worked on and developed a similar system at one point. It worked extremely well. We weren’t an American company and ultimately covid killed us (it was US American orgs that were the most interested in our stuff).
It has some uses, but 95% of what is being used for and 100% of the data centers aren’t it.
Do you think LLMs are being used for this sort of thing? Putting aside the sheer technical mountain of a hurdle that slapping an LLM vision model on top of dozens and dozens of real-time camera streams, the hardware requirements would put the company out of business before they made their first sale.
Computer vision models, which are NOT LLMs, have been around for quite a while now and are very good at doing one thing and one thing only. And they’ll do it well for a miniscule fraction of what it takes to run an LLM.
No, datacentres are not being used for real-time gun detection. The company might have other kinds of infrastructure located in a DC, but not the main video processing hardware.
Do you think LLMs are being used for this sort of thing?
Yes. It took all of five seconds to find out too.
No, datacentres are not being used for real-time gun detection
You’ve already been wrong once, care to try for two?
Using a LLM for detecting a specific object on an image is possible but stupid: if your object is always the same (like in this case) it’s several orders of magnitude cheaper to train once on that specific object then use the computer vision model running directly on the local server that’s recording the video.
Otherwise:
- the api costs would be colossal, 0.001$ per each image, at 30 fps it’s $100 per hour, nobody would pay that
- The detection latency would be several seconds vs almost instant
- Without internet connection the system wouldn’t work
Use cases for LLM-based image recognition is if the object changes at every request or it’s ultra specific with brands and colors
if your object is always the same (like in this case)
It isn’t the same though. A large gauge shotgun and a small gauge pistol are pretty different looking. Compare those to a .22 rifle with a scope, and those to a decked out ar15. That’s a lot of different always the sames. What if it’s a revolver? Or has a folded stock? Or a sawed off stock? Will it recognize a derringer or a mac10 with a large capacity mag as guns?
We can because they make us dead. We have valid reason to fear them which is a great motivator for most species to learn to recognize the danger. You’d still recognize a ring gun as a gun, without getting specifically trained to do so a machine will identify it as jewelry.
so, train the computer vision model for a gun and train again for a shotgun. Run the two detection models at the same time.
Your approach is the typical “but if you really want you can use an atomic bomb to kill mosquitoes” - yes, you could do that, but nobody is paying $1 mil/year in inference costs (+some expensively licensed software to wrap around that) when it can be done locally with a $300 GPU (+ some expensively licensed software to wrap around that)
