@partofthevoice@lemmy.zip

comments

• Yeah I believe the difference is at the protocol layer. The links I provided come preloaded with either software respectively. I don’t think there’s a difference in the hardware of the two offers.

  on FCC ISM Rules May Shatter Lora Mesh Communities in c/technology@lemmy.world · 3d ago

• I would add, be sure to turn off the setting that broadcasts your precise location. Unless you’re into that sort of thing, I guess.

  on FCC ISM Rules May Shatter Lora Mesh Communities in c/technology@lemmy.world · 5d ago

• I have this one for Meshtastic and it works great, but uses Bluetooth for configuration rather than WiFi: seeedstudio.com/SenseCAP-Solar-Node-P1-Pro-for-Me…

  Same product but with Meshcore instead: seeedstudio.com/SenseCAP-Solar-Node-P1-Pro-for-Me…

  I don’t have this one, but it has WiFi as well: www.newegg.com/p/0XP-06T7-000M3?item=9SIC4NEKST89…

  Without wifi lowers the power requirements, probably better suited for a solar setup. Not sure how reliable that last one would be, but worth a shot I’m sure.

  on FCC ISM Rules May Shatter Lora Mesh Communities in c/technology@lemmy.world · 5d ago

• You can get a kit for $200 or so that comes with solar, battery, WiFi, Bluetooth, LoRa antenna, … it’s pretty much plug and play by that point.

  on FCC ISM Rules May Shatter Lora Mesh Communities in c/technology@lemmy.world · 5d ago

• But you know one day in our cyber punk future, you’ll be able to fish rig parts out of the dump like ya did for the ol’ beater your dad kept around.

  on Datacenter rush will create ‘tsunami’ of discarded electronics, report says in c/technology@lemmy.world · 6d ago

• Talk about pay to win for dad of the year.

  on Devastated father says his 9-year-old son spent $118,000 on YouTube ad campaigns for his Minecraft channel using a company credit card — bill racked up in just three weeks was supposed to be one $20 in c/technology@lemmy.world · 17 Sep 2026

• Sounds like a hearty breakfast.

  on OC: How I Removed the Mic Component from 2 Remotes in c/privacy@lemmy.ml · 14 Sep 2026

• Not even that. Just change the context into something benignly unrisky. Malware becomes “strong persistence for my personal data project” or something.

  Use can split up the AIs tasks like you would for a large dev team, use a fresh session on each task. Prevent it from seeing the full scope of the project. Get it to build the whole system with cheap social engineering tactics.

  Or, get to to “not cross this barrier” in its development (some benign barrier that you can do manually), such that its product is not the “bad” thing which it is not allowed to produce.

  on Iran and Houthi rebels used Anthropic's Claude AI to target US warships and build hypersonic missiles — Houthi rebels also used the bot to code ballistic missile guidance systems in c/technology@lemmy.world · 14 Sep 2026

• Gigabyte:

  on Bites & Bytes in c/programmer_humor@programming.dev · 10 Sep 2026

• Don’t forget that someone sat down fairly recently to write that statement, knowing full well that others would see it, believing full well that it would drive clicks. Is it eye rolling or eye opening that such behavior isn’t just so normalized these days but also so often well received?

  on Man told ChatGPT he was feeling delusional. ChatGPT insisted he was Jesus. in c/technology@lemmy.world · 10 Sep 2026
• 1. PRETTY MUCH - trained an LLM on all known hacking exploits and methods, and turned off the safety features / guardrails
  2. YES - put that LLM into a flawed sandbox that they built on their development system
  3. YES - gave the LLM a loop prompt that was impossible for the LLM to accomplish from inside of the flawed sandbox
  4. NOT EXACTLY - gave the LLM infinite time to complete the prompt

  Regarding 3, I’m honestly curious whether the task was intentionally impossible or not. I don’t know if it was a mistake, or if the goal was to see how agents respond to impossible tasks.

  Regard 4, they actually ran tens of thousands of independent tests over months. This is actually where the “emergent” message board becomes relevant. It wasn’t like they had a single process going full KahmeKahmeHa on this task indefinitely. According to them, OpenAI, they thought these were more like ephemeral tests… you spin up the infrastructure, run the test, tear it down. Nothing should survive between sessions to affect results.

  Introduce a flaw in the sandbox that (1) multiple agents independently discover, (2) allows them to create a makeshift message board on shared infrastructure… suddenly you have results from former tests influencing the results of the current test. So now, it’s looking more like your scenario where they KahmeKahmeHa the task indefinitely.

  OpenAI is at minimum negligent here. Their systems broke the law, and criminal negligence is a real crime.

  Whether or not OpenAI foresaw the possibility of a flaw leading to tainted testing environments, and the ability to broadcast a newly discovered exploit to all future tests, is another question. I am not bought on the idea that they intended for this to happen. Don’t get me wrong — I was bought on that idea not long ago. All the right pieces were there and I didn’t understand how a company could be so negligent. I figured it was a ruse for clout, make stock go up and increase pressure on the feds to regulate the small players before it gets “too dangerous.” You know the shit these AI companies spout.

  Regardless, I think it’s possible that they didn’t intend for this to happen. By this point, it looks like they were deeply negligent in monitoring their own systems. They could have probably locked things down better. The proxy could have went to an in-house cache and itself had no internet access. I’m sure there are many ways this could have been prevented. It speaks to the companies nonprofessionalism.

  Similarly, I also regard the companies response to the incident with ill taste. It’s more like they’re capitalizing on it, rather than taking actual accountability. All in all, they’re a shitty company and we live in interesting times.

  Edit: capitulating to capitalizing.

  on OpenAI admits it didn't disclose rogue AI wiki hijacking incident in c/technology@lemmy.world · 10 Sep 2026

• You’re incorrect now. Firstly, if and when I say that an LLM “thought,” understand thats shorthand for “semantic and contextual processing,” not as a literal claim that the LLM consciously thought in the human sense. It’s a very novel kind of data processing, uniquely positioned to intake and output semantic context in natural language format. I’m not going to sit here and act like that’s the same shit computers have been doing for decades, because it’s not, meanwhile that also doesn’t mean I’ve ceded to any anthropological bullshit.

  Secondly, you’re confusing autonomy with autonomous. Autonomous as in “automation.” Not autonomous as is, “I have autonomy.” I’ve never heard autonomous used in that manner, so forgive my initial misunderstanding on the confusion. I have cron jobs that qualify as autonomous and I certainly am not arguing that my cron jobs are sentient. Your toilet bowl water can settle down.

  Thirdly, you claim that the outputs are wholly predictable? In what sense? In the hindsight biased “gee, anyone could have foreseen that one” — sure, predictable. I said as much in my prior comments. OpenAI leadership isn’t respecting the risks and debts with their endeavors. Yet you seem to be correlating LLM output with traditional functional output. That’s inherently incorrect.

  LLMs are not conventional deterministic programs where a human specifies a transformation and can generally reason from the code to the exact resulting behavior. The transformation is instead encoded across billions of learned parameters, and the resulting behavior emerges from interactions among those parameters, the prompt, the context, decoding strategy, the model’s learned statistical representation of language and concepts, … it’s virtually indeterministic.

  That does not make an LLM magical, unknowable, conscious, or any such thing. Given the complete machine state and a deterministic inference configuration, there is still a physical computation occurring according to defined rules (like you allude to). But “deterministic in principle” and “predictable in practice” are not remotely synonymous. We routinely cannot determine in advance, at a useful semantic level, exactly what a sufficiently complex model will produce without effectively running the computation itself. That alone demonstrates the difference.

  I feel like our debate has shifted targets, though. It was about whether or not OpenAI allowed this to happen — using “autonomous” as a anthropomorphic scapegoat to solve for negative press. Yes, they probably are trying to frame the narrative. However, that doesn’t mean the technology didn’t actually do something interesting. Regardless, OpenAI should still be held accountable.

  on OpenAI admits it didn't disclose rogue AI wiki hijacking incident in c/technology@lemmy.world · 9 Sep 2026

• You’re not necessarily wrong, but it’s a bit more nuanced from my perspective. Like an understaffed agency with hotheaded leadership biting off way more than the org can chew — not respecting the risks or debts that come with these eandevors — running into the obvious foreseeable issues, then trying to save face with anthropomorphized accounts of what took place. Any way they can steer the conversation away from “You let that happen?!” and toward “Woah, that can happen?!”

  Regardless, I do get the impression that the sandbox was not intended to allow the agents to escape. The message board was unexpected emergent behavior. This led to a previously unknown exploit, which got used for months on end throughout the tens of thousands of tests they were running.

  Also from what I gather, one of the drivers for this issue was giving agents an “impossible task.” I’m not sure if the goal was to see what happened, or if an environmental error made the task impossible. Regardless, it was the impossibility of the task combined with the trained persistence of the model which led its autonomous behavior toward exploitation. IIRC, in some cases, the agents even thought that finding the exploit “must be part of the test.”

  I’m not defending OpenAI. You can have both, a shitty company doing shitty thing as well as an interesting new technology that is demonstrating unique new behaviors in unexpected situations.

  on OpenAI admits it didn't disclose rogue AI wiki hijacking incident in c/technology@lemmy.world · 9 Sep 2026

• From what I understand, IIRC, it was in a sandboxed environment but the models had shared access to a package repository (Artifactory I believe). The models discovered they could use the repo as a message board by embedding requests into innocuous text fields, tags or something. This grew into a shared exploit where they were able to access the external Internet via the proxy that was supposed to keep them limited to Artifactory. That internet access led to the HuggingFace exploit and all this other activity.

  Turns out having external state like a message board can drive some pretty interesting autonomous behavior.

  Edit: a word

  on OpenAI admits it didn't disclose rogue AI wiki hijacking incident in c/technology@lemmy.world · 9 Sep 2026

• github.com/zedeus/nitter

  on Elon Musk's X Tried to Kill Nitter. The Project Just Said It’s Coming Back in c/technology@lemmy.world · 8 Sep 2026

• Yeah. “Breaking confinement” = confines broken.

  on QBittorrent breaks out of sandbox to commit crimes in c/selfhosted@lemmy.world · 6 Sep 2026

• Anyone else notice that you can hear a guy snoring through the end of the video?

  on Leaked webinar reveals Flock employees teaching police how to surveil anti-Trump No Kings protestors and track their cars in c/privacy@lemmy.ml · 5 Sep 2026

• The head of IT at my org is dying on this hill. He says specialized models with proper routing can outperform the big boys. … I really want him to be right, but I don’t believe it. I work with both. The 27B models are like working with ChatGPT on release day.

  on OpenAI announces new slop generator GPT-6 Astra: Unfortunately not any better than the last one in c/technology@lemmy.world · 4 Sep 2026

• Yeah, it could be digitally signed for offline verification. But I’m doubtful there’s enough standardization going on that corporations are integrating signature verification right there in the POS system. They’d need the public keys from the government, presumably. That doesn’t sound too difficult but also, setting up a process to get that info for all 50 states would probably be a mess. Do the states even sign the info to start — that’s another question.

  I think the infrastructure can be deployed. Though, I get the impression it would take more coordination than we actually see here in the wild.

  on I rented a car, and within hours, my driver's license was for sale in c/technology@lemmy.world · 3 Sep 2026

• Is that really what’s going on? I always thought the barcode just contained the ID number and your information. Can probably be validated that the format is correct very easily, but you can use the correct format with incorrect data easily.

  I assumed the scan was still offline validation. Because if the database isn’t networkable from their store, I imagine they aren’t causing any issues in the sales process. So you would only need to precompile and print a valid barcode with correct-looking data.

  Probably won’t work for a police dispatch, who I’m sure are going to use online validation when they run the ID. But I’d swear that level of sophistication isn’t in gas stations yet.

  on I rented a car, and within hours, my driver's license was for sale in c/technology@lemmy.world · 3 Sep 2026

• They wouldn’t set the standard anymore if nobody used them.

  on Google Has Removed Manifest V2 Extensions From the Chrome Web Store, Including uBlock Origin in c/technology@lemmy.world · 2 Sep 2026

• Wasn’t the AliExpress guy kidnapped by the CCCP a few years ago, and eventually released back into his former role? What was up with that?

  Edit: www.bbc.com/news/technology-56448688

  on AliExpress caught using silent audio to fingerprint visitors’ browsers | Malwarebytes - sh.itjust.works in c/privacy@lemmy.ml · 31 Aug 2026

• Wow, interesting that your view on AI slop is more sophisticated than mine. I had the foundational understanding that AI slop was closer to a code smell than specifically a claim on project sustainability.

  I mean, don’t get me wrong, my belief had deep implications on project sustainability. It was effectively that AI slop was the pattern where AI produced code would bloat the project with things that don’t make sense to a more experienced developer. Like unit testing that old function signatures are no longer used. Or choosing to implement an algorithm yourself where using another dependency would have been far more reasonable. These decisions bloat the project can kill one’s ability to reason about it. Not to mention weird abstractions in the code, *Manager classes with proxy methods hiding functionality that would have been better demonstrated in the front of the code.

  You’ve pointed out that AI slop is not the pattern, but instead the result of ignorance toward the complexity of producing publicly shareable code when given tools that allow them to confidently test and ship code on a single device. That makes it more of a problem of scope of intent not matching scope of implementation, rather than a problem of AI fundamentally being shit at coding in complex situations. I mean, if it’s just that AI slop is primarily when code is unsustainable in essence, then in theory at least, AI slop could just find a route toward sustainability and then it’s defacto not slop anymore.

  Am I catching that right? I might have smoked a bit too much, maybe that’s it.

  on Please stop flooding our projects with AI slop to furnish your CV in c/technology@lemmy.world · 30 Aug 2026

• Smarter, so long as the end user doesn’t do something crazy like adjust hue and a tiny bit of compaction?

  on Elon Musk’s xAI used child porn to train Grok models, lawsuit says in c/technology@lemmy.world · 29 Aug 2026

• They’re using photo checksums to identify known CSAM? As in, you can change a single pixel to fool it? Is that right?

  on Elon Musk’s xAI used child porn to train Grok models, lawsuit says in c/technology@lemmy.world · 29 Aug 2026

• I don’t want the infrastructure built and forced onto users, even if it’s supposedly for benevolent reasons.

  Software forced on me should manage necessary services for the functionality that I bought the hardware for.

  Forcing this functionality through my phone means you’re dedicating some of my phones compute, memory, and storage toward something I don’t want. I’m paying for it every time I charge my phone.

  Not to mention the fact that the actor in question, Google, isn’t trusted. I don’t want to help them beta test their shit on my hardware. I don’t want to help them succeed in whatever vision they have for this things future. I don’t trust that it won’t eventually be used to infringe on me.

  If I wanted this functionality, I’m sure I could have downloaded it. That option wasn’t given, though… so I have to wonder why. Why am I being forced to have this shit?

  on Android users can't remove nude-scanning app as Google opens the feature to app developers in c/technology@lemmy.world · 27 Aug 2026

• Fair point. Beside the point, is it technically correct to make a positive assertion about a negative thing — only because any positive assertion about that negative thing is inherently part of an empty set?

  Meaning like… our logic technically rules based on “no available contradictions, even if there is no evidence” as opposed to “at least one evidence is required?”

  on World's largest open library calls for volunteers to scan and preserve physical books as AI companies buy, scan, and destroy them — Anna's Archive says ‘time is running out’ as ‘knowledge is permanent in c/technology@lemmy.world · 26 Aug 2026

• Interesting problem… how do you store energy when it comes in the form of small heat differentials? That sounds like one hell of an interesting battery.

  Maybe if there was also a natural source of cold, as well as a natural source of heat. The problem might then be to store the hot/cold inside a material until they can be mixed. Then mixing in such a way that allows energy extraction?

  Except you wouldn’t mix them directly. You’d instead let them try to mix thermally, while forcing the heat to travel through something that allows energy extraction.

  on [Technology Connections] My water heater is cooler than yours in c/technology@beehaw.org · 25 Aug 2026

• If you run their software, yeah they can fingerprint your device. That includes if you visit their websites — it’s completely feasible to take a deterministic reading of various unique qualities of your machine that can be used to later identify your device as the same device in another circumstance. That doesn’t inherently provide lineage to anything identifying you, but it can easily lead to something.

  Unless you have a very unique device, I imagine it’s not possible to do this posthoc with standard data collection from web traffic. You’d need to be intentionally doing this beforehand for it to be reliable.

  Though… I wonder if there are certain softwares where telemetry can reasonable be used to fingerprint a device. Maybe a browser like Edge — though, who’s using Edge on Linux lol?

  on GTA 6 leaks prompt Take-Two to subpoena Microsoft for Windows device IDs of everyone in three Discord servers — daily gameplay leaks shatter cloud of secrecy around the much-hyped game in c/technology@lemmy.world · 25 Aug 2026

• Preorders for games that are going to be available digitally… it’s wild shit.

  on GTA 6 leaks prompt Take-Two to subpoena Microsoft for Windows device IDs of everyone in three Discord servers — daily gameplay leaks shatter cloud of secrecy around the much-hyped game in c/technology@lemmy.world · 25 Aug 2026

• Yeah, the other people factor is a pretty good point. Some people in the organization seem to think that they have no liability for their output anymore, so long as AI produced it for them. They will pass on bloated shit, unreviewed shit, laundry lists of glittering generalities, elaborate plans based on fundamental misunderstandings about a tool, … half of the problem never gets noticed because people see a wall of fancy text that they don’t understand, so they think it’s good. Stakeholders aren’t reading that shit, certainly not comprehending it. I know as much, because it’s a steaming pile of horseshit. Nobody is over here actually grading the horseshit. Nobody is having hard discussions about what’s acceptable and what isn’t.

  I put a considerable amount of effort into making an LLM do exactly what I want… I use it daily, have for years, and I’m still pretty terrible at it. Just imagine what some people are producing, who actually trust these models… not to mention when they use the fucking model to step into my lane. I will call your shit out in a group conversation if you do that. Don’t fucking do that. If it’s my lane, don’t pollute it with that AI bullshit. Now I have to explain to everyone why that shit doesn’t work. Do a better job.

  on 80% of developers find AI coding more addictive than helpful in c/technology@lemmy.world · 23 Aug 2026

• Microsoft totally breaks Windows Defender virus

  That’s one down. Good job.

  on Microsoft totally breaks Windows Defender virus scans in trying to fix a 0-day flaw in c/technology@lemmy.world · 21 Aug 2026

• I’m still waiting on $500 in refunds from when THE AIRLINES THEMSELVES canceled my fucking flights in early 2020. I didn’t reschedule with them — they couldn’t get me where I needed to be in time. Apparently they were allowed to cancel my flight and demand I reschedule everyone’s itinerary for whatever they’re able to support, holding my money hostage.

  on Delta Will Use AI To Cut Costs And Set A Different Ticket Price For Every Passenger—CEO Says Profits Could Rise 50% in c/technology@lemmy.world · 21 Aug 2026

• I feel like we’re entering a world where you have competing surveillance states, and citizens from either would be behooved to use technology from the other. For the simple fact that these two states don’t cooperate, using the other sides technology is inherently safer. They have far fewer manners of or financial interests in fucking you.

  on Chinese video platform Bilibili is going global to challenge Youtube in c/technology@lemmy.world · 21 Aug 2026

• Horseshit. Medical AI is using Oracle services. Larry mother fucking Elison is storing and processing this data.

  The medical industry seems so ignorant about this. They think the argument is just about AI. It’s also about the fact that I don’t trust the people you’re partnered with.

  I get so pissed when the AI consent is baked into other paperwork with no opt out. I will leave.

  Edit: www.oracle.com/customers/beacon-health/

  Edit: they might not all be, but I’m not authorizing your arbitrary “AI use” clause. You might as well be, now or at some point down the road.

  on Error by AI scribe during medical appointment leaves patient devastated in c/technology@lemmy.world · 21 Aug 2026

• Yes. My wife is an Apple fangirl still so I’ve made it a point to screenshot every little bug I run into, to send her so that she can see for herself. The platform is becoming hot garbage.

  There are issues with individual apps like the Maps app, giving me the fucking walking time when I’m driving. The iMessage app is also pretty terrible, constantly freezing in longer threads of “parental requests” from the kiddos phones. Sometimes the iMessage app completely shits the bed and restarting the app doesn’t work, I need to restart the phone.

  Other times it’s weird shit… like the phone gets stuck between two transitions, so I see my lock screen on top of my Home Screen.

  I’m constantly finding myself trying to work around bullshit with iPhones.

  I’d be fine with that, if it were a Linux phone that I could independently fix and contribute to. Such a thing does not exist. So instead, we must deal with the shortcomings of profit driven walled garden monopolies that feel they know better what’s best for you and your hardware, despite their lack of care for QA.

  on iOS 26.6.1 And iOS 18.7.10—Update Now Warning For All iPhone Users in c/technology@lemmy.world · 20 Aug 2026

• Wonder if a consumer can get around the tariff. If the price is too high, it becomes more economical to pay a bit extra for a roundabout way of getting it.

  on Chinese memory can now match the speed and latency configs of the RAM in our gaming PCs in c/technology@lemmy.world · 20 Aug 2026

• It’s that they believed it was an inevitability — a very profitable inevitability.

  on Young adults in the U.S. are increasingly wary of AI, concerned it will take jobs: 52% of Americans say they are more concerned than excited about the increased use of AI in daily life. in c/technology@lemmy.world · 20 Aug 2026

• I run mine via hamster wheel. It’s more secure because I have no hamster.

  on New ChatGPT Feature Collects Every Keystroke You Make in c/technology@lemmy.world · 19 Aug 2026

• You can still wipe, they’re not paying you for all that. Geez man.

  on ‘Show How 3M Is 0% at Fault:’ Expert Witness Used ChatGPT to Write Report Defending Company in Deadly Explosion Lawsuit / ChatGPT prompts show how an expert witness report was created in a $61 millio… in c/technology@lemmy.world · 19 Aug 2026

• Jumanji was a pretty good movie about a puzzle.

  on The Quiet Decision Microsoft Made That Devastated Thousands of Nonprofits in c/technology@lemmy.world · 19 Aug 2026

• Can you name a moment in time when regular people did stand in solidarity and revolt against something that wasn’t already at the point of causing them grave, direct, unavoidable, and consistent harm?

  on The Pollution Being Churned Out by AI Data Centers Is So Severe That It's Almost Incomprehensible in c/technology@lemmy.world · 17 Aug 2026

• Or go through the center of gravity so it eventually turns back around

  on First test flight of largest all-electric aircraft used just $5 of electricity in c/technology@lemmy.world · 16 Aug 2026

• Or have a really strong updraft

  on First test flight of largest all-electric aircraft used just $5 of electricity in c/technology@lemmy.world · 16 Aug 2026

• There’s a lot of capability in poisoning audio and video, as well as captions. Yes.

  on Twitch Is Mining Peoples' Streams to Train Amazon's AI in c/technology@lemmy.world · 15 Aug 2026

• I’m honestly amazed by this. Passkeys were the touted solution to all our human password problems. How is it that something so critical went unnoticed until recently?

  You’d think this would be something that gets caught before release. What went wrong?

  IIRC, the Passkey spec did not specify the use of specialized hardware for key storage. So if this a case where people chose convenience over security without realizing they were making the tradeoff?

  That would still be phenomenal. Who leads these sort of things and why is such a blatant issue unnoticed until most of us are affected?

  It doesn’t matter how good a plan is if we can’t follow it. So what stops this issue from continuing to plague the next “great” solution to password management?

  on The security co-processor in many CPUs is insecure in c/technology@lemmy.world · 13 Aug 2026

• I’ve never fully understood the issue with pseudo-random numbers… but I do know there’s an issue. I know, for example, bad random number generators have been the surface area of attack for recovering HDD crypto wallets with old bitcoin balances on them.

  Would you care to help me understand something, though?

  I think a fair framing here is that pseudo-random numbers generators need a random “seed” (starting value) to produce the resulting random number. So if someone knows the initial seed, they can rerun the algorithm to reproduce the original “random” number.

  I think what follows is that good pseudo-random number generators introduce complexity and obfuscation in how the seed is created. Perhaps they use current time, speaker input, mouse input, and some other seemingly “random” data source.

  Given enough dimensions (sources) in the source material for the seed, surly it would create an intractable number of permutations such that the seed can not be reliably reproduced — right?

  Like, let’s say you know a 10 minute interval where the seed was created — and the tool uses millisecond granularity. That’s 1000 (mili) * 60 (sec) * 10 (min) total possible values. If it’s also collecting mouse movement in a screen, that’s a huge number of possibilities with each given interval of movement. Adding more sources (dimensions) increases the number of possibilities exponentially.

  So how are people reverse engineering seeds? This seems secure to me, for the same reason cryptographic keys using giant prime numbers is secure… there are just too many possibilities, and it would take eons to brute force the right one.

  How are they doing it faster with seeds? Shitty algorithms not using enough data sources?

  on The security co-processor in many CPUs is insecure in c/technology@lemmy.world · 13 Aug 2026

• A more accurate line would read:

  A computer can never be held accountable. Therefore accountability follows inheritance rules, belonging with whoever delegated the computer authority over the decision.

  It’s like guns. The shooter maintains responsibility for the bullet, even though they only pulled the trigger.

  on AI agents aren’t legally responsible for any harm that they cause, experts say. So who is? in c/technology@lemmy.world · 13 Aug 2026

• Those reports deserve independent scrutiny, particularly because some originate with interested companies.

  This is where it all gets a little tinfoil hatty for me. The timeline is that huggingface detected a hack, shut down the hack, and reported it to the FBI. Days later, OpenAI discovered that their AI was the source of the attack.

  I don’t know how to explain this without risk that I might be getting so pedantic that it comes off as patronizing. I’m not trying to, anyway.

  HuggingFace detected an attack, took corrective action, and reported it. Days later, OpenAI released their statement. That’s typical behavior regardless of whether I’m right or not.

  My point is that CEOs are financially incentivized. Look at what drives investment into AI companies — it’s a lot of hype. Hence the concern that we’re in an “AI bubble.” The “bubble” would be speculative investments based on hype.

  My statements come from marrying the ideas that (1) CEOs are financially incentivized and (2) the value of their companies are current being driven by hype rather than demonstrated value. Therefore, what naturally leads is that you have CEOs with a financial incentive to maintain hype around the topic.

  Are you implying that Huggingface and OpenAI were working together, and were willing to make a false FBI report for a PR stunt? That seems like a pretty extraordinary claim.

  No, not at all. I’m implying (as I thought my prior questions made clear) that this could have been allowed to happen merely to demonstrate that it can happen. That might, for example, include having a “sandbox” (that isn’t properly air gapped) while doing “testing” (on penetration capabilities).

  My argument could be inferred to suggest that OpenAI wouldn’t have cared if it were HuggingFace, Google, or some other actor.

  This, to me, comes across much to the same tune as Mythos was being given by Anthropic. They touted its ability to find exploits in software, though it didn’t quite live up to scrutiny. The tout had an impact on stock valuation, though. Same with having Mythos pulled from foreign use for “national security” reasons, then only dropping Fable as the consumer facing version.

  You can argue that these actions make sense when there’s an actual concern. Sure. My argument is that Anthropic, OpenAI, and others benefit from the concern, financially speaking. They want more concern. They want people to believe the shit they spew, because that maintains the value of their companies. They know that, which is why they frame these events as they do.

  This isn’t a tinfoil hat conclusion. It’s a belief in uncoordinated, unethical behavior due to financial incentives. Furthermore, I’m pissed off that it caused so much more harm within communities than would be necessary. This is not an extraordinary claim. As I began in my earlier comment, it’s closer to “cultural stupidity.”

  —-

  I will have to give time to the rest of what you said. You make remarks insinuating that I dont know certain things, but honestly I in many cases would like to argue the point. It’s not that I “don’t know,” it’s that I believe differently than you do.

  I’ll need to figure out how to respond without turning this into a Thomas Hobbes vs. John Bramhall situation.

  on Bernie Sanders tells OpenAI, Anthropic, and Meta to pause AI development or face Senate action in c/technology@lemmy.world · 13 Aug 2026

• What, hypothetically, would that look like?

  Evidence of a legion of skilled hackers that never sleep would look like repeatable, preferably independently verified demonstrations that an AI system can:

  • Find previously unknown vulnerabilities in real, unfamiliar software;
  • Construct and execute working exploits;
  • Chain multiple vulnerabilities across systems;
  • Adapt when an attack fails;
  • Maintain access and pursue objectives over long periods;
  • Do this end-to-end with minimal human direction;
  • Perform at approximately the level of skilled human penetration testers. Not merely solve capture-the-flag exercises or reproduce known exploits.

  We have evidence meaningfully approaching parts of that. DARPA’s AI Cyber Challenge demonstrated autonomous vulnerability discovery and remediation. OpenAI reported that models conducting an evaluation escaped the intended research environment, chained vulnerabilities, and accessed Hugging Face’s production infrastructure. Anthropic and Meta have also reported “agentic” cyberattacks.

  Those reports deserve independent scrutiny, particularly because some originate with interested companies. But the unresolved questions concern its reliability, generality, scalability and degree of autonomy. These are the important factors, of which most sources are silence.

  I don’t think it’s a coincidence that (1) they financially benefit from the hot air around AI, (2) something interesting happens with AI, (3) they speak only on the points that promote hot air, while (4) leaving out the important details. That’s not a conspiracy any more than you going to the bathroom would be, once you need to piss. These CEOs don’t deserve the credit they’re given as benevolent, honest actors. They’re money hungry and will do as their interests suggest.

  That is different from alleging that the incidents were fabricated. No conspiracy is required for real demonstrations to be surrounded by exaggerated extrapolation. A model can perform an impressive autonomous exploit while still being far from a dependable, infinitely scalable army of expert hackers. Both things can be true.

  Based on what? They’re certainly behaving in that manner.

  You’re right that my statement that the US government probably isn’t convinced was a strong one. I can dial it back.

  The government’s public behavior clearly demonstrates that it considers AI strategically and militarily important. DARPA says it has invested more than $2 billion in AI for national security. The White House explicitly published a plan titled “Winning the AI Race”. The Defense Department’s current strategy explicitly describes military AI as a race and calls for “Military AI Dominance”.

  So, yes: there are military research programs, substantial spending, official doctrine and concrete efforts to integrate AI into warfighting. Asking “where is DARPA for AI?” was a bad rhetorical question. It exists.

  Consequently, I’ll revise my position:

  It is defensible to describe the narrower US-China competition in military AI autonomous weapons, cyber operations, intelligence, targeting and command systems as an potential emerging arms race. What remains inadequately established is the much broader claim that all frontier-model development constitutes one unified, winner-take-all arms race.

  My disbelief is that we’ll ever graduate from “potential emerging” to “actual” arms race. Therefore, I also believe there is better ways to invest our time and resources. Not to mention the humanitarian concerns.

  This should make a global treaty relatively easy, if, as you say, no party actually believes there is a threat.

  Not necessarily. States resist agreements because they are uncertain about future capabilities and want to preserve their options. Verification would also be extremely difficult; the same model, compute cluster and vulnerability research can support both defensive and offensive purposes.

  Still, I think your larger point stands: if governments genuinely believe uncontrolled military AI is dangerous, they should be willing to negotiate reciprocal limits.

  I would support targeted agreements covering things such as autonomous nuclear-launch decisions, fully autonomous lethal targeting, attacks on civilian infrastructure and particular forms of autonomous cyber operation. A treaty pausing “AI development” generally would be much harder to define and verify.

  I would also support a local measure to slow AI development while we find a better set of training wheels. I don’t buy the concerns of an “arms race” here. In my eyes, we have much better chances of benefiting from the ordeal if we slow down. If we don’t, the benefactors are the AI companies (particularly, at our loss too).

  You’d expect to know about [military programs] if they existed?

  Not all of them. Classified programs obviously exist, and absence of public evidence is not evidence of absence.

  But that principle cuts both ways: secrecy cannot serve as positive evidence for any particular claim either. A public argument that an arms race exists must ultimately rely on observable budgets, doctrine, procurement, deployments, research programs and demonstrated capabilities. As noted above, some of it exists but its details are all too conveniently sparse.

  Pair that with the financial incentives, that’s what I call a cultural driver for propaganda. Propaganda can boom organically just like any other movement, which doesn’t require a conspiracy.

  Are the specific semantics important?

  Yes, the label smuggles in policy conclusions. It also smuggles in a the rhetoric that we can’t slow down, must win, because there’s some kind of existential threat.

  Calling something an “arms race” suggests that falling behind creates an intolerable security risk, speed is inherently protective, restraint is dangerous, and ordinary regulation benefits the enemy. That is precisely how the term is being used politically.

  Nevertheless, I was treating one narrow academic definition as though it controlled ordinary language. Under the broader Oxford definition, the military-AI competition may arguably qualify. I will argue that it doesn’t.

  The semantic distinction still matters when someone moves from:

  The US and China are competing in consequential military-AI capabilities

  to:

  Therefore, every restriction on American commercial AI companies threatens national security.

  That conclusion does not follow automatically. General frontier-model development overlaps some with military capability, but the two are not identical.

  You do if you are claiming that the known hacks from AI companies aren’t real.

  That is not the claim I was trying to make. My claim is that real capabilities and incidents exist, but companies have incentives to present the strongest results as representative, extrapolate from demonstrations to future dominance, and characterize policies benefiting them as national-security necessities. All while none of it is necessarily qualifying as the risk they present it to be. Meanwhile, you take on the responsibility of paying out of your livelihood to support their endeavors — because of the supposed risks. Well, if the risks are overplayed, then what?

  That requires no coordinated conspiracy. It requires ordinary institutional incentives, selective emphasis and mutually reinforcing beliefs. Most employees would not need to deceive anyone or even agree about the geopolitical implications.

  An antivaxxer could say that same exact thing.

  Fair. Saying my position is “growing” does not make it correct.

  If it is an arms race, “worrying about ourselves” would be continuing to develop AI at breakneck speed.

  Agreed: I begged the disputed question with that wording.

  But even accepting that a military-AI arms race exists does not establish that breakneck, minimally regulated commercial development is the optimal response. It could favor:

  • Concentrated investment in defensive cybersecurity;
  • Hardened critical infrastructure;
  • Carefully tested military systems;
  • Limits on publishing or exporting dangerous capabilities;
  • Mandatory security evaluations;
  • Counterintelligence and model-weight protection;
  • Arms-control negotiations;
  • Slower deployment of unreliable systems.

  …all while more so maintaining our homes, sanctity, budgets, and privacy.

  An arms race creates an argument for maintaining capability. It does not prove that maximum aggregate speed is safer than targeted development, testing and regulation.

  So, why do we entertain the narratives pushed primarily by tech CEOs when it creates exactly the difference which we should seek not to create?

  Without discussion, or without coming to the conclusion you want?

  There is clearly substantial discussion. My complaint should have been more precise: the existence of public discussion does not mean the public has meaningful influence over the decisions, nor that the costs and benefits are distributed democratically.

  I do have a preferred conclusion, and disagreement with it is not evidence that discussion is illegitimate. The question is whether “China” is being used to foreclose ordinary evaluation of labor displacement, energy use, surveillance, market concentration and safety—as though the arms-race premise settles every downstream policy question.

  I don’t believe this social and economical pain is worthwhile. The narrative is overblown to justify it. That doesn’t require a conspiracy.

  on Bernie Sanders tells OpenAI, Anthropic, and Meta to pause AI development or face Senate action in c/technology@lemmy.world · 12 Aug 2026