Not even remotely true, terrible reporting by Tom's. AI PCB utilization isn't particularly high, cell phones and mobile devices consume an order of magnitude more PCB area in a given year, but that doesn't fit the current "AI bad" meta so any problem in the world is now AI's fault.
What actually happened is that Trump started a war in Iran, Iran sent missiles over to KSA, which blew up the petrochem facility who supplies 70% of the global supply of resins used in PCB manufacturing: https://finance.biggo.com/news/3xNDpZ4BtCxy99G55-nj
Somehow, their wide range of results touch every single area of math except anything touching cryptography. I think it's fair to expect that three letter agencies are given the ability to veto the publication of certain discoveries.
Wasn't there a comment a while ago when all major LLM providers went down almost at the same time that it was caused by a sort of virtual "Room 641A" thing?
Some version of this has been posted a dozen times now. The dog barking and baby crying detection are advertised features of this product range which is used for SmartThings integration.
Swarms of RE-hopefuls are now using tools they don't understand to work on projects they don't understand. Anyone doing RE knows step 1 is read public documentation, FCC filing, patent applications, etc. That step would tell you what the device does out of the box, no RE required to find everything the OP has actually uncovered.
Instead they busted out Claude and have been gaslit by their own AI into thinking they've discovered something. They just discovered features that Samsung has been pitching for years.
I don't get why people think it would be worth all this effort for Samsung to nefariously work out not only if you have a dog and/or a baby, but exactly when they make noise. Like, sure, that would be a gross invasion of privacy if they were doing it - but what's the business reason for putting all this effort into it?
I can almost see the "they're using it for advertising" or "they're selling information about whether you have a dog or a baby, of course!", but notifying every time they make noise?
There's obviously more to this story than these sensational knee-jerk reactions assume. They're just so non-curious for a supposed 'investigation'.
Well, person didn’t have a dog and then they suddenly do, and that will be noted by the device. I don’t understand the downplaying here, even though it’s correct that these people not necessarily know what they’re doing. The device does collect information about you and does listen (all while costing thousands of dollars). And these spy features they announce are not a product for common users but for data companies.
I don't think I'm the one being overly credulous here.
None of the assertions you made are backed by this 'research' which as far as I can see only shows that the TV is capable of detecting dog barks and baby cries.
You're just presuming that because the TV is capable of sensing these things (which Samsung advertises - links at end) this data is being collected and sent to Samsung - and then sold on.
So, some questions:
Is information about every time your baby cries or dog barks even valuable enough that this is plausible?
Is there any evidence that this is actually being sent over a network - for purposes other than the features they advertise?
Is there evidence that these are even being monitored if the dog bark or baby cry feature is not turned on?
Shouldn't the burden of showing things like this be on the people making the sensational accusations?
I could come up with more, but I think that these are enough. I'd love (well, hate, because I don't want it to be true) to see some actual proof of any of this.
Well yes, but when was the last time any police force had the mental ingenuity of asking the Samsung corporation about details on on someone's living environment? And what level of accuracy would you trust that intel?
Yeah the police would never ask Amazon ring for mass video without a warrant either right? Oh wait they could and they did. And it was shut down after it was exposed.
And on flock cameras they still do.
It's like this, once you have the data they will come for it.
"They recently had a baby and now they suddenly don't" is probably a datapoint they sell to ambulance-chasing medical malpractice lawyers and PE-owned grief counseling clinics...
Different people learn new things differently. Folks without a background in RE, who are playing with these tools, may also not care about privacy until they find a reason to. I don’t see anything wrong with folks taking a roundabout way of finding out how things work; Be it software or privacy.
I think the issue is mostly over-sensationalizing findings.
For example, it would be tiresome to have someone insist that s0ss is clearly an illegal hacker with multiple felonies to his name solely because he chats with YCombinator's secret anonymous hacking collective. And yet this post right here makes it very clear you are one of us! ;-)
While caution is definitely warranted after LG, I am definitely getting the impression that people are far too quick to jump to conclusions because an A.I. service told them something without them factually checking it first
If normies understood that baby crying detection also requires everything-except-baby-crying detection, they would be up in arms about being spied upon... and here they are
Sometimes, the essence of a story and end goals has more value than the actual finding and story itself. Most us probably don't know this is part of a documented feature, and this is the first time I'm hearing of SmartThings integration. I'm still not quite sure what the point of these detection are.
Dog barking is, like, 1% of the findings. They also claim the TV is:
1. taking screenshots of your screen
2. fingerpriting devices on your local network
3. connects to bluetooth mics in the room and recording audio without asking etc.
This reads like you want to gate-keep people first exploring and discovering the absurdity of the surveillance state we live in.
In my opinion, Claude "gaslighting" is a signal the other way - If Claude thinks they've discovered something, that means the llm-compression of most digitalized human knowledge still finds SmartThing a disturbing piece of tech that user do not want.
First off, not all gatekeeping is bad. The locked doors at the power plant marked caution are locked because the wrong move will kill you and it will hurt a lot the entire time.
So first we have to the examine the gate being kept, and who's the keeper. If it's "being allowed to vote" and the gate is being kept by "men", sure, tear down that wall. In this case however, the gate being kept is not reading the documentation and being mind blown for something that's documented. Not like, hidden away somewhere, but plainly documented. I'm torn though, because xkcd 1053 also applies. Unfortunately, tech is one of those areas where you can only cry wolf so many times before people just totally ignore you.
> Swarms of RE-hopefuls are now using tools they don't understand to work on projects they don't understand
I mean, that’s just a statement of fact from the grandparent, no? I don’t see it as gatekeeping; are you focusing on some other part of the comment?
I think someone who’s aware of how to do Reverse Engineering being tired of those who’re making mistakes but advertising the output of those mistakes as legitimate discovery is pretty normal. I remember when I was learning FreeBSD that I’d show up in the mailing lists thinking I’d found bugs, but really I hadn’t read the docs well enough, and people were direct with me that I needed to do better if I threw the word “bug” around. It’s just normal peer feedback.
I would respectfully submit that trying to treat Anthropic’s models’ nonsense alarmism like it is legitimate output is closer to gaslighting. At best, the output shows that the majority of people have an alarmed tone.
The comment seems to imply to me that they are too gullible to understand what the LLM tells them and are being gaslit into believing things that smarter researchers would know without needing an LLM. Honestly, the comment seems dismissive of the current reality in general.
I don't see how you get to the conclusion that the original commenter thinks people are being gullible.
I look at the original comment and it doesn't say anything about gullibility. It expresses frustration that there's an influx of people interested in the field of Reverse Engineering but aren't digging deeper to where the answers are. I read the comment as being more similar to the complaints of the Eternal September, when an existing culture got permanently swamped by people who wanted to interact with the tools of the culture but didn't realise there was a process to learn first.
For myself, I'm glad there's an influx of people interested in technology and being able to grasp at the tools that can deliver outcomes. I don't love seeing people appearing that want to tell me that deeper understanding is dead, not because it does me any particular daily harm but because I think there's still a lot of value in understanding the details of a topic which summaries tend to gloss over. I've been feeling this for a long time, though, it's not related to LLMs but to a culture of knowledge more fixated on soundbites than on getting at the truth of things.
In this case, I see the original commenter reflecting my own concerns, but that's so often the way: we see things as we are, not as they are, which may be colouring my read.
Reasonable minds can disagree, of course, and I respect that we may just be reading the comment quite differently.
If actual researchers with real experience are disagreeing with LLMs, I'd generally still trust the real experts.
At a dead minimum, I'd expect something like "addressing the concerns the experts are raising" instead of trying to do a character assassination.
I generally read "character assassinations" as suggesting that someone has no real technical arguments available - conversely, the experts here are citing specific facts and evidence.
Dog bark and baby crying detection are advertised features for SmartThings integrations. Door bell and door knocking are also included, this is surfaced directly if you hook the TV or sound bar into SmartThings.
I had to check this and it's true. Not saying Samsung or 3rd parties don't take advantage of any data they can collect, but the story the tweets tell is super biased against them. Here, the detection seems limited to specific pattens required for a particular feature.
These folks recently did tests on major TV brands, not accepting the on-device consent forms, and then recording how much data they continued to send after consent was denied (which itself is not always possible). https://www.youtube.com/watch?v=t_GK06_3UY8
Samsung was the single brand reviewed which actually stopped sending data after the user told it to do so.
In February Samsung agreed to stop collecting the data when opted out in the US after a few lawsuits were lining up. LG was the target of a different lawsuit in Texas which they settled and then wrote up a forced arbitration clause and doubled-down on data collection, even when opted out.
Oh don't worry, shitty WebOS is enough to put me off that (I unfortunately own one). However in this case we were talking specifically about that recent study of the effects of opt-outs, and for LG TVs the opt-out did result in the no longer detecting any tracking.
Still, calling it "PRIVACY HORROR" as per original post is typical twitter-style sensation seeking. It's an opt-in dog barking detection that causes local chip to listen for... dog barking
At this point everyone knows you have a cat, dog or baby. Facebook, Instagram can scan your photos and see it. Same with Google via Photos' latest privacy changes. Google also knows if you ever get receipts emailed to you. Your credit card company and bank know, the CC company resells that data too. Shit, Keurig knows because their smart coffee maker scans the network and sees your baby monitor or smart pet food dispenser. All these companies are rediscovering and selling the same basic facts about you and your household.
If they wanted to know if you had dogs or babies and wanted to spy on you, they'd just transcribe the audio. People that have dogs or babies talk about them. Going to the effort to detect these and advertise it as a feature so you can secretly use it for marketing doesn't make a lot of sense.
I will add: Use the TV as a HUB for all IOT, light sensor so you know when the light is on or someone passes in front of the TV and there are some others but I can't remember now. They are all advertised in the Smart Things app.
I hate 2 things about my samsung tv's (I have 4 only samsung): the simplistic remote and ads on home screen. Even with custom dns sometimes they get through.
Enterprise* customers are overwhelmingly more likely to primarily host on Azure than GCP, m365 runs the corporate world top to bottom. OpenAI doesn't care where the customers come from, more platforms == more reach, they have zero reason to be preferential to any single hosting platform.
* By "enterprise" here I mean "big company that is not a tech company". If most of your userbase is carrying a mac, you're probably a tech company. If you're mostly making widgets in factories, chances are high Windows and the ecosystem around it is involved.
Android does the same MAC rotation that Apple does, both on WiFi and Bluetooth. Working with IoT sniffing solutions at home, I've found it's reasonably easy to detect when a single MAC is randomly changed: it'll be broadcasting its presence, suddenly go dark, and then another mac with same first 3 bytes starts broadcasting with a very similar RSSI. This works with BLE adverts pretty well and allows me to filter out MAC noise from household devices while I'm trying to sniff something on my bench.
Some version of this claim has been made for the past 4 years. There's a data cliff, there's no more compute to buy, the financials don't make sense and all of these orgs will be out of business by end of quarter.
Not once has any of these predictions come true, the pace of progress has continued on it's exponential trajectory since ChatGPT first came to the public's attention.
So why now? What is special about today that suggests all of this is coming to a screeching halt despite all evidence to the contrary?
> ... the pace of progress has continued on it's exponential trajectory since ChatGPT first came to the public's attention.
Did it? Model wise? I would understand agents wise, sure. But model wise? The attention to detail from the model? The ability to recall minute things? Improvements are there, yes, but mostly on Fable and Astra. Opus still isn't as attentive as Fable in long term writing for example.
Sure, Opus 5.5 benchmarks better than Fable. Sure. But is that the model, or is that the RL for agentic work?
From where I'm standing, the model work has not been exponential at all, and more and more it looks like the latest and greatest is getting too expensive too fast. Both 5.5 and 5.6 chat models got nerfed, actually nerfed not the tea leaves kind. In mid 5.5 cycle the chat model lost the ability to substitute names if given an outline. 5.6 cycle the chat model lost the ability to use paragraphs after a few hundred words (coinciding with Chat/Work split).
There's a race from OpenAI to serve dumber models on chat. I'm not even sure who they are racing against, but the fact that Astra, Sol 6.0, and now Sol 6.1 not being available for chat, should tell you that those models are expensive, and not the kind of models that can be freely "chatted" with on a subscription. OpenAI much prefers you use Work and limit the chat usage, much like Grok and Claude. I'm guessing they will announce that later during the dev days.
That could be cost cutting too, true, but really? That's the only explanation? And nothing else?
Sure, the progress did not stop. But it is nowhere near close being exponential when it comes to LLMs themselves. Agents are separate.
tl;dr it changes the weights, it does not add new ones.
RL makes the model better within its capabilities, it does not increase the total ceiling of the model. Ie does not make it smarter. Qwen 3.8 27B is a great model, still probably not at the limit of 27B in terms of coding capabilities, and it still has that "small model feel" to it. The better smaller models get at coding the worse they get at everything else too.
Going from Sol 5.6 to Astra, Opus to Fable, you can still get that "larger model feeling," though less so. The bigger models can reference things that you would not have expected.
The distinction I'm making is that models themselves are getting too expensive, so the improvements are mainly on the RL side. Which is fine, but they do not make the model smarter, rather make them use their capabilities better. They are likely to catch things they are RL'd for, and that hopefully anything else doesn't get negatively affected. RL'ing for Javascript world for example did not improve the C world when working with the models.
Hmm interesting idea. I’m pretty confident there is generalization and learning that occurs during RL that does make the model smarter. So I think the distinction doesn’t fully hold up.
Qwen 3.8 27b is not smarter than other 27b models. Smarter, as in its ability to recognize minute yet important facts has not changed. If you ask it a for a code sample it produces a better sample, true, but it has not been able to surpass that small model feeling.
For 27b model, it works tremendously well in agenic tasks too. It generates stupid amount of tokens even for the simplest tasks and gets feedback from the harness to eventually produce something right.
I would not call that the model got smarter. It is better at coding, but it still cannot recognize subtleties that frontier models would catch first try almost 100% of the time. And yet some benchmarks show Qwen 3.8 27b is at Opus 4.6 levels.
This is why I differentiate. Grok 4.5 and 4.6 is the same base model with the latter being a post-training refresh. Same thing for Gemini 3.7 Flash and 3.8 Flash. Some people say that for certain 5.x era GPT models. Again, improvements are there, but the base models are same/similar, and the model is just able to display its capabilities better.
Is that smarter? In a certain sense yes, in a certain sense no. I would say it is moving to the model's local maximum, and bigger models are still smarter, even if they are not able to display it.
Grok 4.7 is a good example, the model is bigger, has more attention to detail, but the post-training is botched somehow and it is worse at agentic tasks. Is the model stupider? Or is the agent stupider?
I didn't use the word LLM. I'm talking AI capability, you're focused on this or that current approach to AI. I think it's fair to assume that the approach will change as new ideas are learned, new and more hardware will be purchased and applied to the problem, and then capabilities will (for now) continue on their exponential curve, same as it has gone for the past several years.
These things are knocking down Millennium Prize problems while a substantial subset of commenters here are still thinking about stochastic parrots.
>Not once has any of these predictions come true, the pace of progress has continued on it's exponential trajectory since ChatGPT first came to the public's attention.
I think we'll eventually hit an information theoretic type of wall with physical hardware and GPUs and need a similar AI breakthrough as well as the development refinement of logical/physical qubits in the quantum computing space with some analogue to the transformer architecture to continue accelerating. However, I think there must be many years of development and refinement that can take place before that paradigm shift to overcome the physical compute wall is necessary. This is just my theory, but I'm young enough that I'm expecting with the rate that we are advancing, I will see AI / LLM analogues developed and run on a quantum computer in my lifetime.
Has it been exponential this whole time? I feel like GPT-4 was pretty dang good. Maybe it’s rose tinted glasses cause I could finally have a bot write my dockerfiles and bash scripts, which knocked my socks off
Yep. I've made the claim (and been wrong). I was convinced the data cliff was going to be a real problem. Now I feel like we are on the cusp of having Tony Stark's Jarvis at our fingertips.
Incredible how many times I read similar comments over the years, containing 'on the cusp' and 'what a time to be alive'. Indeed, what a time - not a single user-facing thing on the internet has improved since then, considering the power tool we got. The most used web services get drowned in generated stuff and so are the users
Not a single thing? In my house, we are using LLMs to:
- plan youth soccer practices
- develop well-formatted soccer game substitution schedules
- build and ship software in languages I haven't used in 25 years on platforms I've never programmed for
- do meal planning and build shopping lists
- prepare grocery shopping carts
- solicit medical advice
- perform Garmin watch data analysis
- administer devices (with SSH access) using natural language
- avoid counterfeit soccer jersey purchases
- create "Warrior Cat" graphic novels
- make cartoon strips
- troubleshoot appliances
- manage finances
- review accounting ledgers
- diagnose malware infections
- so much more
And we do it all from a simple prompt that we can talk to if we choose.
I've built more (and better) software in the past month than I did in any given year in the 30+ years I've been programming.
I can understand pessimism regarding how this affects society. I can understand pessimism regarding how this gets abused. But for the life of me there's no good reason at all to be pessimistic about how quickly this has improved.
> I've built more (and better) software in the past month than I did in any given year in the 30+ years I've been programming
I feel similarly, but I think it's a valid question. Why is all the software I'm using not getting better? To be honest, I feel it's more buggy than it's ever been.
Companies need to radically change to be able to take advantage. Most companies are afraid to do that and are letting their engineers serve as slow meat proxies, doing software development basically the same way as before.
The difference now is that they've hit the "good enough" point. LLMs are a tool, and that tool is useful but not incredibly valuable unto itself.
To make a manufacturing analogy - ChatGPT was a manual machining mill, and in the years after we've gone from that to a 3-axis CNC mill. Now we've added a 4th and 5th axis, which is great for the 2% of parts that need that functionality. But the big win was that initial jump from manual control to CNC. Why would I pay an extra $2 million for my CNC machine when I could just design my parts to be simpler to produce instead? The AI labs are trying to make these incredibly complex tools, but the market doesn't want/need them so they're competing on price for the tools that people do use. By selling their metaphorical CNC machines for half of what they cost to produce.
Oh, and we've bet the entire economy on the hope that fancier CNC machines will magically solve all our problems in all industries, from healthcare to the legal system.
So - will AI progress continue to improve? Sure. Will we continue lighting money on fire in order to make it happen? That remains to be seen.
>The difference now is that they've hit the "good enough" point.
In some aspects sure, but in others no. Open AI's goal is to build "highly autonomous systems that outperform humans at most economically valuable work." and Astra was a big jump in that. There still isn't a better model for computer use and vision/spatial work. Driving, Operating Robots, Video Editing, 3D modelling, graphics are all things Astra was >>> at than any other model. I'm sure you don't care about any of that so it's easy enough to slip by you but this analogy - "Now we've added a 4th and 5th axis, which is great for the 2% of parts that need that functionality." is dead wrong.
Right, that's exactly my point though. Those are absolutely valuable use cases. But are they useful enough to justify a trillion dollar valuation? Or is most of the economic value in the stuff that already exists, and can be performed nearly as well by qwen/deepseek/kimi/GLM/etc?
And beyond that - how long until those individual Astra capabilities are distilled into separate Qwen-27b size models, with harnesses and scaffolds specifically designed to support that functionality?
>Right, that's exactly my point though. Those are absolutely valuable use cases. But are they useful enough to justify a trillion dollar valuation?
Replacing white collar work would be worth dozens of trillions of dollars at minimum. Software is not the only valuable job that can be done on a computer.
OpenAI and Anthropic already have what it takes right now to become trillion dollar companies even if the above doesn't materialize.
Chatgpt is used by a billion people every week. Their ads program hit $1B Annual Revenue Run Rate in 200 days. And Anthropic is growing so fast they're on pace to hit $100B in Annual Revenue.
>And beyond that - how long until those individual Astra capabilities are distilled into separate Qwen-27b size models, with harnesses and scaffolds specifically designed to support that functionality?
How long until...you could say that about the capabilities of past models but OpenAI still dwarf everyone else in consumer usage, and Anthropic and OpenAI are still growing enterprise usage heavily. In the end, neither the billion+ users of gpt or the enterprise customers are going to give a shit about what qwen does. And specialized models often perform worse than generalized ones.
This is how I feel about it. I've stopped looking at all the scores of new releases and just look at the price to see how much usage I can get in a month. Seems like I'm not the only one either, from comments above like
> "Opus 5.5 is so good that I don't want it to be replaced anytime soon. Stop training models[...]"_
Those points were true at the time and most are still true now. But they aren’t predictions.
- it’s correct there isn’t much fresh data anymore
- it’s correct that compute is scarce, that was 100% the case and a huge issue at the beginning of the year, it is better now but still scarce, and hardware is now way, way more expensive
- it’s correct the finances don’t make sense
But there is no way to know when a bubble pop, because it’s a psychological phenomenon across an extremely complicated distributed system (ie the stock and bonds markets)
The amount of irrationality I see in the economy with AI makes me more convinced that the wall street bankrollers know very well they're throwing money into a pit, but it's a pit they're gambling will turn into some world hunger ending AI (that will somehow also keep them making money off of scarcity)
never mind that theres no guarantee we'll get that mythical AI. Never mind that the societal reformations would also impact their revenue numbers.
I’m pretty convinced Wall Street is clueless and relies mostly on vibes. It’s the same people who thought all SaaS would become unnecessary after seeing a Claude code demo. Though eventually they will have to ask for the ROI, and that’s where the whole thing will calm down
The new hardware (TPU v8 and VR) are more expensive but they are significantly cheaper per flop. e.g. many multiples more performance for only 2x the price.
If I have some ML workload to run I can buy $x of Blackwell chips or I can buy significantly less $ worth of Vera Rubin chips to get the same performance. That's the key thing to keep in mind when you're talking about financials.
I was thinking the same thing in terms of running out of data a few months ago. But aren't most gains in the past year+ due to reinforcement learning in some form? Which doesn't need "fresh data" per se, as the model effectively creates the data as it goes. As long as engineers can come up with proper environments, tasks/goals, rewards, and actions, I don't really see data being a limit to model improvement in an agentic sense. Maybe as a knowledge base
I think this is just a case of the bitter lesson that increasing compute just makes all these predictions meaningless. LLMs just keep going when everyone predicts them to fail constantly.
It’s a shell script. You can download and read it before you run it. Piping it directly to the shell is reckless.
I’m not sure how your machine is configured but mine has permission boundaries and security policies that make sure programs are behaving properly. I don’t run everything with my personal user context.
What actually happened is that Trump started a war in Iran, Iran sent missiles over to KSA, which blew up the petrochem facility who supplies 70% of the global supply of resins used in PCB manufacturing: https://finance.biggo.com/news/3xNDpZ4BtCxy99G55-nj
Nothing whatsoever to do with AI.
reply