Hacker Newsnew | past | comments | ask | show | jobs | submit | FuriouslyAdrift's commentslogin


Problem is that they have wifi chips that will scan and connect to cell phones. The TV manufacturers have agreements with the cell providers for open wifi services on all their phones so you can't block it unless you live in the middle of nowhere and even that may change with LEO satellite service.

I opened up my LG TV because I saw it doing aggressive wifi scanning even when the wifi was "disabled" and physically removed the antenna wire and wrapped the whole guts in RF blocking foil.


I believe you on the aggressive WiFi scanning but I’d have to see concrete proof about the phone part to believe it. Draining users phone batteries via WiFi hotspot functionality that the phone doesn’t even show to the user would be a very big deal.

I imagine if they were going to strike a deal with the phone companies they’d just embed cell chips in the TVs and connect directly.


There used to be several different things for free wifi for subscribers that also got used by national networks. Cellco, AT&Ts home boxes had open wifi that was available to any AT&T subscriber for wi-fi calling, etc.

Comcast/Xfinity does this.

I don't know if they do it either, but if they use hashes they are in no way draining the battery of someone's phone to send a few bytes every so often. That's not me saying it's okay, mind you.

I've told myself if they start putting 5G in TVs I would either glue a faraday mesh to the back or open up the TV and ground out the antenna or remove the WiFi chip if possible.

If they would keep the spying to matching my shows with programs that are being aired or on popular streaming services, sports, or whether I mute the commercials, that wouldn't bother me. Screen grabs would , especially if I use it on my computer. But history has shown this starts off benign and eventually devolve into listening to the room at all hours and keeping it forever in a Bluffdale data center because "think of the children" / "national security".


> if they use hashes they are in no way draining the battery of someone's phone to send a few bytes every so often

I was more thinking of the discoverability aspect. The phone (presumably) would have to be constantly advertising the availability of a hotspot.

But honestly the bigger thing that makes me not believe it is the scale of conspiracy required. All companies, Apple included, would have to secretly be in on it.


> The TV manufacturers have agreements with the cell providers for open wifi services on all their phones so you can't block

That's a strong statement, what's your source?


Where I work, several pieces of equipment from vendors have embedded cellular and wifi that the vendor pays to have service with ATT or Verizon.

That is very, very different to what you described.

I’m sorry what? There are phone network providers that pimp out your phone’s wifi and data connection without telling you?

I’m not sure I believe you? (Not that I can’t believe a company would do something dodgy behind your back, but I just don’t think our phones would go that far off piste - opening a wifi network has battery power implications that I think would have been discovered/widely known before)


Not only that, if you have home internet they will pimp it out to nearby cellphones because that helps patch up weak network connectivity in certain areas.

Comcast / Xfinity definitely does this. I have gotten many 'suspicious activity' CAPTCHAs on popular websites while using my home internet because of this. The routers provided by Xfinity are very locked down so you can't do anything about it.

Let's not even get into WiFi Motion which uses WiFi to track you inside the house. I'm sure it is enabled by default in most households [1]

[1] https://www.xfinity.com/hub/smart-home/wifi-motion


You can get a gigabit router for $100 that works on xfinity which is worth it over their gear. Which is typically sub optimal anyway.

Last I checked you can still turn this feature off.

Please elaborate on that. I want to do that. Was it obvious what to do, and did it not require destroying a bunch of stuff?

I want to do this for my lg too.

I believe they have a bluetooth antenna and a wifi. Was there just an obvious thing to cut?


Found this - looks straightforward, nothing crazy like messing with a board, just look for the wifi module circuit board inside the tv - and remove it - https://www.youtube.com/watch?v=g9WOI5fFtj0

also https://www.ifixit.com/Answers/View/539438/Want+to+remove+43...


Is there any source you can actually cite for the cell phone bit…?

I know the services exist for various things but specific to TVs, it was just a hardware hacking thread on Reddit and a few mentions on X. Nothing concrete.

May I ask what model of TV and if it didn't have a e-key wifi module that could be removed rather than wrapped inside? Mine is older but the service manual seems to indicate there is just a regular module in there I could pop out.

From last year. It has a board at the bottom, but I didn't want to remove it (the TV gave all kinds of warnings and took forever to turn on if it was removed).

> the TV gave all kinds of warnings and took forever to turn on if it was removed

Ugh, I was afraid of that kind of thing. The manual for my older LG indicates there's UI option to disable wifi entirely. Of course, it doesn't actually exist.


Not a shred of proof has ever been provided for something so trivial to test.

Scanning neighboring wifi networks does not mean it's connecting to them.


Global Foundries (fab) is in Germany and Amkor (packaging) is in Portugal

I almost forgot abuot GlobalFoundries and I wonder if they could profitably make DRAM at current prices...

Nope. They don't have the tech.

Only thing that bugged me about Prey is it felt like it was "on rails" and I couldn't just explore. The game is huge and gorgeous...

One of the few modern games that actually got me tense and scared. It is a really good game.


Be aware EDK II has a recent severe CVE to watch out for.

https://app.opencve.io/cve/CVE-2026-97631


That's just in the PXE driver AFAIK.. shouldn't be an issue here.

There are 300 racks of nVidia A100 servers for sale from a refurber right now on eBay...

We're running Kimi 2.8 on a 4x AMD MI300A single server on-premise that cost about $110k.

We will break even from our old Claude spend in about 2 weeks.

It services 30-40 analysts and staff concurrently and is faster than any of the SaaS out there for us.


Wow... I'd like to know more. 40 analysts not spending 100USD/month is 4k/month saved. That means you break even in 27 months. But you probably used Claude much more than that.

Also what harness do you use? Any problems you encountered on the way?


Just one of our analysts was burning around 30-40 million tokens per project.

EDIT: per project, not per month

EDIT 2: digging out an old bill from 8 months ago when we bought the server, $14k per month is what we were running and it was going up, a lot.


Anthropic is very good at either convincing or forcing (I’m not sure which) companies to pay for tokens rather than for subscriptions. My old job moved from subs to tokens and everyone started running out of budget mid month. It was a bad move for a company that laid off all of QA and 2/3 of Engineering and committed to everyone making up for it with LLMs.

I don't get this, why do people pay for API tokens at all? What is the reasoning?

I don't think Anthropic or OpenAI offer subscriptions to enterprises - it's a hobbyist/solo-entrepreneur kind of thing. I seem to recall SpaceXAI had some kind of hack going where they were trying to use subs with Anthropic and then got cut off.

Enterprises also need very stringent IP protections that you don't get through the subscriptions. Even the Anthropic-API per token license isn't great - sometimes you need to go to 3rd parties like AWS Bedrock to get guarantees regarding no multi-tenant GPUs.


You can have subscriptions up to 150 seats with Anthropic.

Ah - so I was somewhat, but not totally wrong. They do offer subs to SMB (I did not know this!) - but for the Enterprise (typically in thousands to tens of thousands) directed through other channels.

Zero Data Retention is the big one I think (and also the ToS don't allow multiple personal accounts for businesses anyways)

I have only ever paid for tokens and never used subscriptions because I need to know that I can generate sufficient real world value at the true cost of inference.

This seems like setting money on fire. Nothing stops you from tracking token use on your subscriptions and tracking how much you would spend via APIs for if/when you can't use a sub any more, and taking the savings in the meantime.

If the stakes aren't real your behaviour won't change

You are forbidden from the subscriptions on enterprise

Claude Teams exists.

No Zero Data Retention there.

They don't have the choice to begin with

I think you should compare with api cost? I am using $200 plan and it is probably around $12k in token cost....

Are $12K worth of token guaranteed or the company may adjust token generation speeds?

Oh boy... we are talking about insane token spends in corps at the moment. 100$ is a blip.

Are they happy with Kimi 2.8's reasoning ability? I haven't tried it.

Kimi is also a lot more willing to do whatever tf you ask it such as vulnerability scanning or PoC development. Or reverse engineering.

Plus on prem can go find abliterated models.

It's an advantage that keeps me coming back for local models when I can't get around Claude or Codex's little blockers.


It's worked out. Definitely not frontier but it's the largest model that works reliably on our hardware. We also try some smaller ones when they come out.

How bad did the electric bill get hit?

We have a data center already and we're in heavy industrial manufacturing. It didn't even register in our usage.

Very cool

There's no Kimi 2.8 as far as open weight is concerned. WTF. This must be a bot and other bots.

If you take the kimi plan you would have seen kimi 2.8 preview alongside kimi 2.7

There was about 6 months ago... lol. We can't run 3.x, it's too big.

The 4x MI300A server has 128GB per card.


Maybe they're talking about parameter count?

TSMC dominates world wide with 80% of the general chip market and nearly 100% in some specific markets.

Due to Trump policies (and geopolitical realities), they are building or helping to build fabs in the US and elsewhere. It's takes years to build a fab.


This reminded me to check about ESMC (joint venture between TSMC and Bosch, Infinion and NXP, but mostly TSMC) since I haven't heard about it in a while. Seems they are on schedule which somewhat surprises me.

More than half of all server CPUs are AMD. They absolutely domininate in HPC and servers (and have for years, now).

2026Q2 numbers show Intel falling quickly but still in the mid-60% to 70% in every market: https://www.tomshardware.com/pc-components/cpus/desktop-cpu-...

AMD was at over 46% of the "server" market 1Q2026.

Intel isn't even condered a player in HPC anymore. 147 of TOP 500 is AMD (it's really custom chips, which AMD also does with it's Xilinx and Instinct divisions, and nVidia that are chewing up the Intels)


AMD has practically abandoned their HPC-focused Threadripper series too, leaving the market segmentunmet by anyone. The supercomputers in the Top 500 listing only account for a fraction of CPU sales, where they mostly serve to feed accelerators.

AMD has an entire division dominating in HPC and #2 close behind nVidia in datacenters with extremely good margins. Over half their revenue is from those segments. Not a lot comes from consumer devices right now. That's why they did so well against Intel right now. Intel is the one that is failing in those markets.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: