What just happened? Another incident has taken place that illustrates the need to be careful what you tell AI. A Florida woman is facing felony charges after she used Claude as a diary and allegedly wrote that she planned to “shoot up” the Sheriff’s office. After a human reviewer examined the statements, they were reported to police.
According to the arrest report, Carli Michelle Heller, of Bonita Springs, Florida, wrote on September 26 that she would attack the Sheriff’s office. She later said that she uses Anthropic’s chatbot like a “diary.”
Claude’s safety systems flagged the entry and it was escalated to a human reviewer. After deciding it was a credible threat, the reviewer reported it to law enforcement.
The company says it may share user information in limited emergencies if it believes disclosure is necessary to prevent death or serious physical injury.
Deputies identified Heller and visited her home. She was detained without incident before an LCSO intelligence detective took over the investigation.
Heller faces a charge of making a written threat of violence under Florida law. Florida Statute 836.10 makes it a second-degree felony to send, post, or transmit a written or electronic record threatening to kill or injure someone, carry out a mass shooting, or commit an act of terrorism. The communication must be made in a manner in which another person may view it.
Anthropic isn’t going to be taking any chances when it comes to anything it deems a potential threat. Last month, it was reported that OpenAI and Sam Altman are being sued by British Columbia over claims that the company could have prevented a mass shooting in the Canadian province.
The shooter, eighteen-year-old former pupil Jesse Van Rootselaar, had previously been flagged by OpenAI’s safety team for her conversations about gun violence, but OpenAI never alerted police because the conversations did not meet the threshold for legal referral.
In June, Florida also sued OpenAI and Altman, alleging that ChatGPT had contributed to real-world harms, including the 2025 Florida State University shooting.
The latest incident is another reminder to think before you enter something into a chatbot that could get you into trouble. It’s certainly not a private diary whose contents are for your eyes only.
Reports last month revealed that human contractors reviewing Microsoft Copilot’s image editor can see users’ prompts, uploaded photos and AI-generated edits. Documents show that some of those assignments contain sexual, disturbing or potentially illegal material, though the reviewers are not there to flag the content – only to assess whether the output is accurate.
OpenAI and Sam Altman are being sued by British Columbia over claims that the company could have prevented a mass shooting in the Canadian province.
Not sure how to feel about this one, on balance I’d say probably fuck British Columbia as well as the rest of them? A government suing someone for not snitching is major Stasi energy, even if mass shootings are horrific.
And people were not-so-long-ago concerned that phones are spying after them. Now they just send all the info themselves.
People will fight you tooth and nail to prevent something they view as immediately threatening to them. On the other hand, they’ll pay you to implement something they view as immediately useful to them. In either case, they don’t care at all about any non-immediate repercussions for anything.
I’m building a fully private assistant with local Home Assistant, local model, and a local Wikipedia database for lookups. Will probably make a post/release of it soon once I finish replacing all my Google Nest displays.
I’d like to subscribe to your newsletter.
I want to do this, but only got as far as setting up a minimal Home Assistant. Only put a few hours into it, though, then got sidetracked by trying to turn an esp32 into a speaker/microphone puck.
Hah, you can watch my site: https://nowsci.com/ as I put everything there.
I’m very nearly at daily usable by everyone, including my partner who has a very different accent to mine. A lot of it was tuning sentences/words to catch poor STT to skip going to an LLM to speed things up and make it reliable. Here’s the current hardware in action:
@[email protected] yes, Ive tried a lot of LLMs and have settled on Gemma 4 2b with a very specific prompt and local MCP tools for unit conversions, weather, lookups, etc.
For Q&A, my current build uses an MCP container I made to do a SearXNG search and top result retrieval for context (I use a trick to weight Wikipedia results). The final step I haven’t completed is for this to be a backup to Kiwix, for which I also haven’t found a great search tool/index/MCP, so if you want to collab on that I’d be interested. One less thing for me to build myself ;)
deleted by creator
deleted by creator
deleted by creator
Up until the 80s, there was still some debate over whether wiretaps should be legal. It’s sad that we have come to accept them as normal.
It was less worrisome back then. The cost to have someone listen to and transcribe your conversations was quite high. They’d really have to want to go after someone to do it.
Now, turning voice conversations into searchable text is nearly free.
Not only that - wiretaps have Judicial Oversight in that they require a Court Order (well, in most of the World, not sure if that’s the case in present day America) which a Judge supposedly only issues if there’s reason to suspect the person whose lines are tapped of committing or preparing to commit a crime, whilst Antrophic’s surveillance of people’s conversations with their chatbot has no Judicial Oversight at all and has no minimum legal threshold that must be reached before it is allowed to happen.
I think this is one of the greatest dangers AI presents. Sure it makes some of us stupider, sure it drinks our water and eats our GPUs and raises power prices… but once upon a time, it took actual human effort to run a dragnet, even with judicial approval.
Now, you plug in an LLM to the giant pile of corporate surveillance data and ask it to find troublemakers. Anonymity is dead and its death made retroactive and since there’s no government oversight, the government can just lease access to the tools they used to need approval to use.
Back in the day the 31337 found a way around the bots.
Too bad they weren’t more elite, or they could have prevented the modern internet from becoming what it is
Where does this apply?
To the English-speaking world and probably beyond.
i know very few people who wereor are concerned about being spied on regardless of how much they should be.
Worse, if you tell them how big of a problem it is, they’ll assume you have something to hide
I get called a downer for mentioning how this all works. You shouldn’t connect your smart tv to the network, you shouldn’t use Alexa, anything you do online is public, and it always gets recorded poorly.
Maybe my partner is the only person who listens, and she has all her stuff disabled, there are no network attached smart things in it house, but she still uses windows and she’s not yet receptive to moving to Linux (which is a hard sell).
a lot of thise things are great though. everyone draws the line differently. im well aware of the surveillance. to live significantly under the radar you’d need to cut out tso much more to make it meaningful. do you really just watch antenna tv and dvds? no ordering online? always use fake membership cards at the grocery? pay cash? i mean fuck. i get ya though. im just playing devils advocate for the people that may not know and are calling you a downer. EDIT: there are so many nice linux dustros now and windows is so bad, that should be one of the easiest sells imo.
No I bought an AppkeTV, which while it isn’t perfect is not doing what smart tvs do.
I’m not living as a hermit or completely off grid. I just stick to stuff that tracks me as little as reasonably possible and am mindful of what information could leak.
Like, I still have to go to hotels when travelling and they have zero data security with your ID photo and cards. I can’t fix that or avoid it. That’s life. But disabling the always on microphones in my home I can manage.
What is a smart tv doing that an AppleTV isn’t?
Automatic content recognition and selling it like LG and Samsung, inserting ads like Roku, installing residential proxies like 80% of smart tv apps.
If I’m running Netflix and they’re selling the analytics data, it doesn’t matter what platform I use. But everything around that can be avoided.
I would like to use a Linux box but haven’t found a good HDMICEC plus low power plus good app support combination (happy to take suggestions though)
At least your partner it the one you most need to have and benefit from listening. Your privacy can easily be compromised by close associates. Best of luck to you; you’re not doing anything wrong. Its not you that’s the downer, it’s the circumstances you’re describing.
Smart TVs just find those “open” hotspots that cable providers leave on everyone’s modem now; you don’t even need to agree to TOS before it starts locating and spying.
Every device is a wiretap. Every device is a wiretap. We live in a panopticon. The purpose of AI is to price us out of the component market. You can’t escape if you can’t afford your own hardware.
I haven’t seen any proof of this, so I’m still skeptical.
I didn’t approve any TOS setting up my tv, so this would be legally a nightmare for them.
Gamers Nexus recently released a 2-hour video on LG smart TVs and how they spy on you. I haven’t watched the entire video yet, but it seems that the damn things start recording data as soon as they receive power.
so this would be legally a nightmare for them
To many corporations such as LG, laws are just the cost of doing business. Assuming they even receive a fine it will be so small that it won’t matter.
I just mean there’s enough tech illiterate people out there that they don’t need that liability.
95% of people will just turn on the tracking and not know or think about it. Connecting to nearby wifi is more of a risk.
I suppose it should be pretty easily verifiable. Just set up a new SSID without protection and see what connects.
Might as well try it with my own TV. IIRC I never accepted any TOS either.
ZDR policies really are a must. If a service doesn’t have it, don’t use it.
ZDR?
They’re just going to be like “Google it, I’m not your Mom!” or something to that effect. This happens a lot here. Someone comes along, drops some niche term and doesn’t bother to explain it at all because they either don’t know how to imagine what it’s like to not know what they know, or they just don’t give a shit about not being understood by people who don’t already know.
You somehow managed to craft a reply that added even less value to the conversation than the one you’re complaining about. Amazing!
Zero Data Retention
deleted by creator
Presumably zero data retention? It seems to me that what you’d want is for it to never be processed off-device in the first place, but I suppose to each their own.
Surprise! No one could have ever predicted that. 🤯
Don’t use “AI” for anything you wouldn’t mind seeing plastered all over the Internet. You wouldn’t tell a human being that you were planning to “shoot up a sheriff’s office” would you? Why would an AI keep that a secret?
doesnt seem like the right charge. not sure she should be charged with anything though. seems like she should get some help and make sure she isnt gonna shoot nothing up and maybe get some learning on how computer shit works, but if this doesnt make her curious then nothing probably will.
a bigger issue is that many if not most devices are headed to having everything ai integrated so even if she diaried in a notepad file it would get scanned and reported.
Yeah, I know someone frustrated with their local council who privately says “I’m going to 💥 the council” as a joke. That’s very different to sending a threatening letter to the council itself. Or harassing them.
The example here feels closer to the former than the latter.
The problem is that America’s answer to every problem in society is jail. You guys gotta come up with more than “arrest bad guys”.
We say this, but every time there’s a mass shooting people go “Why couldn’t this have been prevented?!?”
Like pick one or the other, but let’s stop asking for predictive policing in one breath, and in the next one asking why someone saying she’s going to shoot up a place is being taken seriously and punished.
If she had said the same thing to a psychotherapist, even they would have been obligated to report it. It’s really not surprising.
Sure, get her mental health treatment, and have a professional assess her for threat level. But go ahead and blacklist her from purchasing firearms, and if she already owns any then put them in some kind of stewardship where she can’t access them for a set time period, and after that can only check them out with supervision for specific reasons like “going to the firing range” or “hunting.”
Not saying she should be convicted of a felony, just that it’s not surprising that this would be reported to the authorities.
And I’m against tech surveillance, but it’s not like this conversation was done locally on her device. She transmitted her messages over the internet to a server in a data center where it was processed. Did Anthropic make any promises that messages were kept encrypted in a no-access vault where even human reviewers couldn’t access them? Cause I don’t think they did…
Not the person you’re replying to, but I think there’s a difference between claiming she wasn’t charged with the right thing vs saying no one should be charged for planning a crime. I don’t think the intention of the law was to prevent people from writing things wherein they thought they were just venting privately. I understand that her understanding was incorrect, but I think the intention (venting vs creating fear and chaos) is an important factor. I think the law is to prevent posting bomb or shooting threats online or sending them directly to people.
I think there should be legal difference between saying “I hate X so much I just want to kill him” and “what gun would be best to kill a person of Y height and Z weight and what would be the best way to hide their body.” I’m familiar with the chat logs from the FSU situation and it wasn’t just venting. I’d probably want to see her full logs before making a legally binding decision, but from the summary it does not sound like a credible threat was made. It may have constituted a referral to law enforcement, but based on what I read law enforcement should have just done a check in and likely not charged. They are likely just salty and want to punish people who aren’t waving thin blue line flags.
I see your point, but isn’t that determination for the court to make after reviewing the chat logs and consulting with competent professionals?
Not that I have much faith in the florida state courts to do the right thing, but honestly if we’re gonna say that a person directly claiming they want to shoot a place up doesn’t call for some form of intervention, then we need to stop clutching our pearls every time a mass shooting happens.
Note: I’m not expressing a personal opinion that it would be pearl-clutching to ask those sorts of questions in the aftermath of a mass shooting. I’m just pointing out the parallels to this situation and the logical inconsistency in saying “well this one was caught early, so it shouldn’t have been; but those ones that weren’t caught in time should have been!”
Like pick one or the other.
But also, fixing society and its many systemic issues, improving the quality and availability of mental health services (and reducing the stigma and increasing access), and ceasing to hand out guns like candy would probably be the most effective ways of preventing mass shootings. But it won’t be 100% even if we can achieve all that (and that’s a big if), so taking threats seriously should still be a part of that strategy either way…
slavers need more bodies
She clearly didn’t intent to act on it and didn’t intend for it to be read by anyone else than her AI boyfriend. So it’s not matching the criteria of the law. Legally, she did nothing wrong.
Why would you even use Claude as a fuckin diary?
First 2 words of the title answer that question.
Primally for the voice part. You record yourself on a drive or something.
Not saying I ever would, I can just see how someone might get there. Especially since it’s a popular yt video format.
You can do that with the notes app too
Well, I think it’s both the reaction and transcription that someone might want. Does notes do voice transcription now?
And yes, overall, it’s a silly thing to do. Im just guessing how anyone might get to a place doing that.
after a human reviewer examined the statements
What, do they not trust their own AI? The AI they’re screaming keeps “breaking containment”?
Exactly, because that’s all BS. They’re letting it lose on purpose to test a skill set.
Also, there are already report buttons the user can click to flag incorrect output, why is a human reviewer involved before a flag is made?
Human in the loop is kind of a big thing in more reasonable approaches to AI. This shows the limitations of the human obviously but the point is that when making impacting decisions it should be under human responsibility.
Apparently not. Although, it could also be a matter of someone being responsible for a decision like that. If they allow the AI to report users straight to the police, who will be held responsible when things go wrong?
What if you’re writing a fictional story and the AI can’t tell the difference between fiction and actual plans to do something illegal? Just to be safe, it sends the cops after you, because your story has some scary words. Better not write anything about murders or violence I guess.
Sure, but what’s the value of their AI if it still needs human oversight? If it can’t take responsibility, then you need a human.
Gotta say: kinda hurts their sales pitch, IMO. I mean, AI isn’t even replacing humans at Anthropic…
LLMs can’t be trusted to do anything serious, but they can do some boring low stakes tasks. That’s not what their marketing material says, but it was written and approved by liars, so who cares.
You lost me at writing a story and Ai.
You can come up with an idea, and you can write the first draft. Then, you need to specify what kind of processing you expect the LLM to do. Could be punctuation polishing, refining the structure, expanding, suggesting improvements etc. If you’re really lazy, you’ll write a 5 line summary, and let the LLM improvise an entire scene based on that alone.
It’s up to you how you want to use this tool. If you let an LLM write entire sections, you’ll begin to notice a familiar tone emerge. Better keep it on a very short leash.
Again, you lost me here. It’s like saying “I made this pizza, my recipe is I ordered it via Uber. I had to specify my ingredients!”
The difference is that you can adjust your involvement to your liking. The role of an LLM in your writing process is what you want it to be. It could be as large or small as you consider appropriate.
Think of it like writing by hand vs. using a computer with good writing software. You can’t really reorganise your manuscript if it’s written on paper, now can you? That would require rewriting the whole thing. Also, spell checking takes a lot of time, and you probably won’t even notice your grammar quirks. Computers make writing easier and faster. An LLM is another tool, just like the simple editing features we’ve taken for granted for the past 30 years. Ask again in 20 years and people won’t even think of writing without an LLM, just like writing a report by hand is completely absurd today.
I’m old enough to use pen and paper for my calendar, to-do lists, diary, etc. But if my kids want to use a PC for such things when they grow up, I’ll either get them an air-gapped machine, or at the very least have them use my personal server for document sharing.
Stupid, stupid, stupid human. Well, those are the consequences of what she chose to do. Which is write a threat. Threats are not protected speech, or so the courts have decided. We created these circumstances, so we can either navigate them with care or run afoul and be chewed up and shat out by the institutional systems we ourselves built.
sigh.
it’s exhausting, really.
Writing a threat is not illegal. Hell people vent all the time, threatening to do things they know damn well they never will. The law says transmitting a threat to another person is illegal because it can be used as intimidation or conspiracy. I think there’s a damn good argument that she didn’t knowingly transmit it to anyone. She thought she was writing in confidence to a non-person. And if it’s not her first time expressing outrage to AI, she would know it would only try to deescalate her.
Who knows how the case will go. I can see it going either way — probably depending on how much money she has. Laws are going to have to catch up to technology in this case, and it’s going to be bumpy. I definitely agree with erring on the side of caution. Private, local AI is far safer, but of course it’s less capable for now.
The article very much states that it is in fact illegal to write down a threat under Florida law.
I think this is a combination of the law being poorly written and the woman in question making a significantly questionable decision. Her actions have consequences (even if I think that law is problematic at best). There’s no reason for her to have had an expectation that Claude chats would be private (we’ve had repeated news reports that prove otherwise).
The article very much states that it is in fact illegal to write down a threat under Florida law.
I don’t know that that’s my read here. The emphasis on the below quote from the article is mine.
Heller faces a charge of making a written threat of violence under Florida law. Florida Statute 836.10 makes it a second-degree felony to send, post, or transmit a written or electronic record threatening to kill or injure someone, carry out a mass shooting, or commit an act of terrorism. The communication must be made in a manner in which another person may view it.
It sounds like it’s not illegal to write the threat down, but it’s illegal to share it with others. If you do keep a private journal, or write your feelings onto slips of paper and burn them (idk if this is actually common, but it’s always been recommended to me after a breakup), that’s not illegal, even if you describe wanting to do something illegal. It’s the sort of hair splitting that wouldn’t normally be relevant, but could make a huge difference here.
Of course, she probably will be legally considered to have sent, posted, and transmitted it, so she’s SOL on that end, but she could credibly (imo) claim she was unaware that another person might view it.
She did transmit such a threat though. Which is against the law quoted in the article.
Yes but not in a way she would think another person may view it, most likely. Obviously IANAL but surely intent matters here?
If people actively realised someone can read all their chats, I’m sure these services would be far less popular.
And here’s the autism slop because I can’t think of the proper term for “actively realising” and do not have the energy to make it readable anymore:
What I mean is, I’m sure plenty of people know in the back of their mind that their data is not protected there, but it’s not the first, or second, or probably even tenth thing they think about when starting a chat.https://vger.to/lemmy.world/comment/26171736
This is a link to my original comment on the subject. I did not say what you think I said in the comment you replied to or the original comment I linked. Her ignorance doesn’t absolve her of her actions or the consequences. Intent does matter to some extent but there have also been a lot of news reports about how data you share with corporate LLM’S isn’t private and so then the question becomes should she have known and not just if she did or didn’t know.
If someone else steals something and sells it to you or gifts it to you and you had no knowledge that it was stolen you can still be prosecuted under the law for receiving stolen goods. Your intent in the matter should be considered because there is a difference between knowingly receiving something you know is stolen and not reporting it, and just getting something without knowing it’s legal provenance and not reporting it. But you can still be charged either way.
In this case though an attempt was made to notify the user that her chats were not private in a EULA or TOS. They changed that TOS relatively recently (something like a year ago I think). At that time it looks like users would have been required to agree to an updated TOS with specific verbage about the sharing of chats that doesn’t look like it would have required users to read the entire TOS, just the agree popup. I linked the article for that below because while that was for allowing AI to train on user data, it should have been a clue that the chats were not private.
https://www.bgr.com/1953473/anthropic-train-claude-chats-how-opt-out/
https://www.anthropic.com/news/updates-to-our-consumer-terms
I appreciate the links! I still don’t think it’s a slam dunk either way, because even in Anthropic’s release, it never actually says that a human will be involved. “Training” could be anything, and while intent does not always matter, a threat is one of the few crimes where it really does.
Yeah half of the online gaming world would be in jail if this was consistently applied… intent must matter in a sane world.
That’s why I said she was probably SOL on that side, but the other requirement for it to be considered an illegal threat is less cut and dry
The article very much states that it is in fact illegal to write down a threat under Florida law.
With the intent that it may be seen by another person. That’s not the same. It’s not illegal to write a threat in your diary.
That being said the summary doesn’t include the fact that the following day she told Claude she had bought a gun and this was their “last chance,” and only then did Anthropic call the cops.
That seems exactly the sort of situation folks have been criticizing AI companies for not summoning help, and makes it seem far less likely that they called the cops over angry venting, but an actual intent to harm.
I’ve long worried that the government will steal the perfect pour over recipes me and the bot are cooking up.
I don’t wanna get Epsteined over good coffee
… But a man can’t live on his knees either
AI and idiots go together like peanut butter and jelly.
In my school district, kids kept getting called to the office because of what they wrote in Google Docs and their search queries. Sexual reproductive scientific research was also impossible to do
The communication must be made in a manner in which another person may view it.
So she shouldn’t even get charged.
Another human viewed it it though.
Yeah, I’d wager that is what the case will hinge on. She wasn’t aware anyone could view it. Otherwise it’s incredibly unlikely that she would have made such a statement to Claude at all.
On the other hand though, I’m sure Claude’s parent company and the prosecution will argue that she didn’t have an expectation of privacy where this statement is concerned (given TOS/EULA, in combination with the number of news articles/reports on the subject).
i think anthopic was right to report it, but the cops likely overstepped. if she had “threatened” the president she would have been assesed by ss but likely not arrested. i wonder if the same cops she was “threatening” were the ones she was reported to and arrested by.
This is just ass coverage for the company. They’re not going to pursue it. They just didn’t want to get sued for not reporting it if she did commit a mass shooting. The prosecution though might.
I’d argue this is positive coverage for Anthropic. They rightfully reported the incident. End of story. That’s where Anthropic’s influence on the situation started and ended.
It’s a criminal case, not civil. The state of Flordia is pursuing the case with absolutely zero decision making from Anthropic. It’s important to understand how the justice system works, and Anthropic has no more right or ability to influence this case than Moleskine would if she had written the threat in one of their notebooks instead.
