Glad to see everyone agrees this is

1. funny cos they’re crying over stealing what they stole

2. acknowledges this means the weights are actually open sourced (which is how it fuckin should be)

also discussion i’ve seen elsewhere:

3. when considering the energy footprint of chatgpt, also consider the energy footprint of running the internet for 30 years to accumulate all that data they stole. therefore the most ecological option is to extract the weights and then opensource it.

just want to add

4. if the accusations aren’t true (still a possibility), oai is probably deliberately buying time/stock recovery by keeping this discussion in the news rather than everyone discussing how much they suck

5. if large entities are going to capture and then open source each others proprietary weights, that may actually be one of the best outcomes for global humanity amidst this “AI” craze

fpr MP games the server code ¹ should be released to the community when finally taking the servers offline

¹ (or at least binaries with minimum standards for support/documentation)

happy to get into into these subtopics, but it’s also possible i may not be understanding you properly because i agree with alot of what you just said.

what are you attributing the close to 0 probability to?

if you wanna say “whats the probability that CMG was at least partly talking out their arse about their capabilities (and especially any claim they were currently in possession of that capability)?”

i’d also give it like >90% probability they (CMG) are full of shit. in which case you could say i agree with you (to within say 10% error margin).

if you’re instead saying the probability is ~100% that audio surveillance capability cannot possibly currently exist outside TLAs because “someone would’ve published it already” then i really cannot agree. (and afaict that ars article does not support that stance either)

extremely good question to ask OP.

thinking on it right now, perhaps Moon (2009)

Not disputing the three letter agencies

The capability they were claiming to have would make a three letter agency very excited.

sorry i didn’t understand. didn’t you say you don’t doubt TLAs likely already have this capability?

oppressive regimes

most (all?) of whom are operating outside typical legal constraints and likely already have access to the million dollar exploit trade which already exists.

further, i’m not sure how this changes the landscape anyway? its not without precedent that variations on capabilities can be useful to more than one market segment concurrently?

trivial to discover and flag as malware

can you explain further what you mean by this? i’m not sure there’s anything trivial about conclusive analysis of the deep complexities and dependencies of modern smart devices

Apple and Google would also be very keen to find and squash whatever loophole let’s them record without showing the notification.

historically we’ve seen google can take over half a decade to address such things, afaict (welcome correction on this) apple’s generally been faster to respond, and i do agree apple’s current public image attire would be contrary to be seen to enable this. [not simping for apple btw, just stating that part of their brand currently seems to be invested in this]

in reality there are a confluence of many agendas and there’s likely ALOT of global users running non-bleeding edge or other variations on the myriad of sub-system components, regardless of what upstream entities like google implement. suggesting certainty of knowledge on everything which happens downstream without additional evidence appears unsubstantiated imo.

which if true would have been exposed/validated by security researchers long ago.

i agree the probability of discovery increases over time. and the landscape is growing more hostile to such activities. yet i’m not aware that a current lack of published discovery is actual proof it’s never happened.

tbh we have our doubts this leak is directly connected to solid proof “they are listening”.

but we’re not currently aware of any substantiated reasons to say with certainty “they’re absolutely not listening”

well they’re an ad company, so being full of shit is pretty much mandatory.

but i’m not aware of any evidence they’re actually 100% full of shit on this exact issue or not? can you explain a little more how you know for certain they’re full of shit. or you just meant “they’re most likely full of shit”?

can you please explain further what you mean? it could be interpreted a number of different ways.

i’m not sure if this is your point or not? but there is obviously overlap between each of those groups, there’s black sixpack dads, and poor/middle class lgbqti etc etc

anyway imo none of this revived division appears organic. there’s always going to be the odd biggot, but afaict the majority of modern biggots are being indoctrinated and radicalised by an organised media effort (and our leaders are either complicit or ‘inexplicably’ powerless at protecting us from it). for sure these radicalised biggots should do better, but we’re also talking about average people going up against billion dollar propaganda machinery. it’s certainly asymmetrical warfare.

governments have all shifted more to the right on average

it appears to be the case. though afaict none of it appears to be organic.

100% shoegaze fits the bill perfectly for OPs request. i’d add slowdive to that list

there’s some great, highly independent shoegaze on yt

also, some Dandy Warhols & Brian Jonestown probably fits the bill, eg. this track

the PR and lawsuit risk

what risk? facebook & others conducted illegal human experiments. this is an enormous crime and was widely reported yet all fb had to do from a pr perspective was apologise.

as we all know, fb even interfered with with the electoral process of arguably the world’s most powerful nation, and all they had to do was some rebranding to meta and it’s business as usual. this is exactly how powerful these organisations are. go up against a global superpower & all you need to do is change your business name??? they don’t face justice the same way anyone else would, therefore we cannot assess the risk for them as we would another entity - and they know it.

So, while i personally disagree for above reasons, I can accept in your opinion they wouldn’t take the legal risk.

simpler metrics are enough

when has ‘enough’ ever satisfied these entities? we merely need to observe the rate of evolution of various surveillance methods, online, in our devices, in shopping centers to see ‘enough’ is never enough. its always increasing, and at an alarming rate.

local processing of the mic data into topics that then get sent to their servers is more concerning is not much more feasible

sorry i didn’t quite understand, are you saying its not feasible or it is feasible? from the way the sentence started i thought you were going to say it could be, but then you said ‘not much more feasible’?

Voice data isn’t

voice conversations are near-universally prized in surveillance & intelligence. There hasn’t been any convincing argument for any generalised exception to that.

I am not sure they could write it off as a bug

it’s already been written off as a bug. i didn’t follow that story indefinitely but i’m not aware of even a modest fine being paid in relation to the above story. if it can accidentally transcribe and send your conversations to your contact list without your knowledge or consent (literally already happened - with impunity(?)), they can 1000% “accidentally” send it to some ‘debug’ server somewhere.

Are they actually doing it? It ofc remains to be seen. Imo the fallout if it was revealed would roughly look like this

• A few people would say “no shit”
• Most people would parrot the “ive done nothing wrong so i don’t care” line.
• A few powerless people would be upset.

If they truly wanted to have mic access, they could for a long time

agreed

and it would have been known

are you sure?

The reality is it is too expensive

imo this commonly repeated view has never been substantiated.

we’ve yet to see a technical explanation for why it’s “impossible/too expensive” which addresses the modern realities of efficient voice codecs, even rudimentary signal processing and modern speech-to-text network models.

and risky

how so? previously invasive features are simply written off as “a bug”. they barely even need to issue some b̶r̶i̶b̶e̶s̶ fines (typical corporate solution to getting caught), that is the level we’re currently at:

“whoops it was a bug, we’ll switch it off”

“whoops another update switched it on again” (if caught, months/years later)

“whoops some other opt-in surveillance switched itself on again, just another bug ¯_(ツ)_/¯”

as long as they have deniability as a bug, there’s almost zero repercussions and thus virtually zero risk. that is perhaps why a company out and talking about it openly is such a no-no. discussing intent makes ‘bug’ deniability more difficult.

in my experience when reading past the “they’re not listening” headlines, and into the actual technical reports, noone has been able to conclusively rule it out. if you know of conclusive documentation, please post.

then there’s the “they have enough data already” argument. which is entirely without foundation, as we all know very well: nothing is ever enough for these pathologically greedy entities. ‘enough’ simply isn’t in their vocabulary. we all know this already.

[i didn’t downvote you btw]

no idea why you’re being downvoted.

you’re entirely correct, and i don’t interpret you as defending telegram’s lack of user protections

(noticing a real uptick in reddit hivemind around lemmy lately, it’s depressing tbh)

Everything that’s encrypted is generally more private and secure than the equivalent which isn’t encrypted

honestly i think this is due to unplanned voice calls essentially being broken technology now.

imagine we had 2020s email spammers while mail servers had 1990s spam filters, that’s basically where we’re at now with unplanned voice.

They gave a fuck about negative pr

sorry to hear about the stockholm syndrome

The planet does though.

Is that the same planet we currently live on? And borrow atoms from to make our physical bodies?

Poisoning the planet is poisoning ourselves.

Quibbling over which part is direct vs slightly less direct is of questionable value.

And it’s essentially misleading, since carbon dioxide:

1. Is a pollutant and
2. The vast, vast majority of it comes from those same pollution sources which are directly poisoning us.

[https://naei.beis.gov.uk/overview/pollutants?pollutant_id=2].

Notice how public discourse goes round & round in a lively show, but never seems to get anywhere?

This is strawman public discourse, and its largely by design.

Stop thinking, worrying and especially talking about climate change.

Instead talk about pollution & contamination

Everyone can see it. It can’t be denied or handwaved or debated away.

STOP POISONING OUR AIR, WATER AND SOIL.

WE NEED THEM TO BREATH, LIVE AND GROW OUR FOOD. (duh)

When you work in an industry where the entire collaborative workflow of everyone is based on software that doesn’t run on Linux, then not running that software is equal to not being able to work in that industry.

there’s no denying that’s true, though ofc it has alot to do with microsofts very agreessive and anti-competitive practices.

though its all a bit tangential, the main issue i think comes down to what someone means when they say “everything”. certainly if someone said “you can do everything”, i’d expect them to qualify what is (should be) obviously a slight exaggeration as parlance. they don’t literally mean “everything” they just mean most everyday things. i think its fairly common in everyday speech for someone to be able to work out thats what they meant.

in the few rare cases when someone literally means absolutely everything, then yes that silly statement would be incorrect. and if strictly intended with that meaning would certainly qualify as misinformation.

Not sure if when people say you can “do everything that windows does”, they should be interpreted to mean “every single piece of software/drivers ever written for windows was also written for linux”.

By that argument not even windows can do everything windows can do, as there’s plenty of cases of software/drivers no longer being supported.

aye exactly. i’d rather see it gone tbh, but since voting is apparently here, and if we try to work within it, such as mentioned above where hackernews prevents downvoting replies to you.

some other ideas

• permit upvoting but downvotes require a textbox reply (imo downvoting without a valid reason is just noise, and we want signal over noise right?)

• self posts not being upvoted (all posts start at 0)

• no voting until you ‘earn your stripes’. not perfect, but somewhat helps at keeping voting within domain expertise.

eg. i ‘fucking love science’, but just because an answer feels nice to me on nuclear rocket surgery doesn’t mean my vote should count.