RISC-V takes first step toward international standardization as ISO/IEC JTC1 grants PAS Submitter status
cross-posted from: lemmy.ml/post/38267171
RISC-V is an industry standard, like USB or Wi-Fi. The specifications are publicly available under the Creative Commons license and every engineer, wherever they are in the world, can use them to design their products locally, while engaging with the global RISC-V ecosystem.This standard is defined by RISC-V International and its members. Decisions are voted upon collectively, ensuring every member is heard. It’s a model that has worked for us for many years, ensuring any updates to the RISC-V ISA happen transparently, without breaking existing designs, and always in service of the broader ecosystem.
The RISC-V ISA is already an industry standard and the next step is impartial recognition from a trusted international organization.
Today, I’m excited to announce that we have taken that first step. RISC-V International has been approved as a recognized PAS (that’s publicly available specification) Submitter by the ISO/IEC Joint Technical Committee (JTC 1).
This means we’re able to submit draft international papers, starting with the The RISC-V Instruction Set Manual, for consideration as true, international standards.
RISC-V Takes First Step Toward International Standardization as ISO/IEC JTC1 Grants PAS Submitter Status - RISC-V International
At RISC-V Summit North America 2025, Andrea Gallo, CEO RISC-V International, and Phil Wennblom, Chair of the ISO/IEC Joint Technical Committee (JTC 1)., announced that RISC-V International has been approved as a PAS Submitter by the ISO/IEC JTC1.Andrea Gallo (RISC-V)
Perplexity.ai is offering a full year of free AI access
If you’re into AI tools or use ChatGPT often, you should definitely check out pplx.ai/hsnqndt86289
.
They’re currently offering a free one-year trial 🎁
Why it’s worth trying:
Cites real sources for every answer (with direct links)
Fast, clean, and easy-to-use interface
Handles complex questions with context and accuracy
Great for research, study, or everyday learning
Try it here: pplx.ai/hsnqndt86289
What do you think — could this be a real alternative to ChatGPT? 🤔
reshared this
Technology reshared this.
Wrist-Cut Transformation Subculture ✡ Menhera-chan - Capitolo 1
La storia di Menhera-chan inizia con degli istanti banalmente tristi. Rincorsa per strada e subito acchiappata da 3 bulle sue compagne di classe...
After police used Flock cameras to accuse a Denver woman of theft, she had to prove her own innocence
After police used Flock cameras to accuse a Denver woman of theft, she had to prove her own innocence
Chrisanna Elser spent days collecting evidence, from apps on her phone to dashcam footage in her vehicle, to prove her whereaboutsOlivia Prentzel (The Colorado Sun)
like this
adhocfungus, Rozaŭtuno, frustrated_phagocytosis, aramis87, Gikiski, IAmLamp, SuiXi3D e RaoulDuke85 like this.
reshared this
Technology reshared this.
“You know we have cameras in that town. You can’t get a breath of fresh air in or out of that place without us knowing,” Milliman said to Elser, according to Ring doorbell footage of the Sept. 27 encounter viewed by The Colorado Sun.
And he saw nothing wrong with that.
ICE’s forced face scans to verify citizens is unconstitutional, lawmakers say
Videos show ICE conducting random face scans on US streets.Ashley Belanger (Ars Technica)
like this
aramis87 e MyTurtleSwimsUpsideDown like this.
And yet still, somewhere out there, there is a fake or brain dead leftist spouting on about how democrats support genocide.
"Chemo makes me sick, so Ill stick with Cancer"
know the kind of people you’re actually talking about. There is still value in electing the lesser evil, and pushing to get better and more progressive Dems in office (that are usually better at pushing back against fascism anyways)
This is exactly my point. The democrats have huge AIPAC backing and support some awful things, but they are fucking saints compared to the only other options in this political system.
They are the only potential vehicles for long term change and stability exactly the way you described.
Progressive candidates have to be winning primaries despite swimming upstream, and democrats have to continue winning federally despite the bad taste (chemo) they put in your mouth.
Lesser evil politics gets you fascism.
The timeline for when you feel a fascist nation's boot comes down to what your skin color is and where you're from.
Lesser evil politics without pushing for better candidates and also doing community building gets you fascism. Just voting for the lesser evil and calling it a day is technically better than voting for the fascist, but not a whole lot as it lets the status quo slip.
Also, what the fuck else are you supposed to do? Be accelerationist and vote for the fascism? Just refuse to participate and and let whatever happens happen? Both of those sound like ways to also get fascism.
In case you haven't noticed, the system in place now in the US became what it is today under both Republican and Democrat Administrations.
One has to be a tribalist useful idiot to deny that "their side" has done as much to create a Surveillance State as the "other" side - amongst those few things which have bipartisan support in the US are strengthening of police powers and erosion of privacy.
The comparison with most of Europe (with notable exceptions such as Britain and Russia) is very telling: it absolutely is possible to have low crime without reckless invasion of privacy, widespread civil society surveillance, draconian police powers and a pay-to-play Judicial System.
This bothsiderism is pretty thoughtless.
It is true that both contribute to a surveillance state but to equate both is to just ignore all policy differences, actions and more to pretend to be nuanced while painting everything as the same shade of grey, which is a downgrade to even black and white thinking.
This is Politics, it's not 1D or 2D, it's N-Dimensional (with a very, very large N): it's not just possible but pretty much a Mathematical certainty than in a country were there are only 2 parties they will match perfectly on some dimensions, even whilst not at all matching in others.
Trying to dismiss away that aspect of Reality (which is incoveninent for tribalists) with sloganeering like "bothsiderism" is just parroting propaganda meant for simpletons who see reality as having just one dimension where there is nothing more than 2 sides.
It's pretty evident by their actual policies that strengthenning of police powers and the surveillance state are things in which both sides of the power duopoly in the US agree in the most, and it the face of both of those parties being shit on that domain your "yeah, but " discourse is really just trying to distract away from the most nasty aspects of both of those taking big fat dumps on the face of every American, by talking about subtle details in the shape and consistency of each one's shit.
Now, if you favorite party did start to diverge in that, you would have reason to celebrate, but it ain't hapenning and discourse such as yours makes it even harder that it will ever happen - why would the tribe's leadership change their ways when there's a veritable army of tribalist peons going "yeah, but, bothsiderism" at any criticism of what they do, even those parts which are undeniably shit.
This is Politics, it’s not 1D or 2D, it’s N-Dimensional
This is the point I made and that your comment ignored.
it’s not just possible but pretty much a Mathematical certainty than in a country were there are only 2 parties they will match perfectly on some dimensions, even whilst not at all matching in others.
This is a strawman. No person is claiming they don't have any aligning opinions.
Trying to dismiss away that aspect of Reality (which is incoveninent for tribalists) with sloganeering like “bothsiderism” is just parroting propaganda meant for simpletons who see reality as having just one dimension where there is nothing more than 2 sides.
This is you continuing to argue against the strawman.
The rest is also that.
You own post:
This bothsiderism is pretty thoughtless.
Your post starts with a sloganeering, hyper-reductive take of what I wrote.
As I wrote in response, "This is Politics, it’s not 1D or 2D"!
It is true that both contribute to a surveillance state but to equate both is to just ignore all policy differences, actions and more to pretend to be nuanced while painting everything as the same shade of grey, which is a downgrade to even black and white thinking.
In case you're unware of it, two forests can be the same kind of forest even when the trees in each are different: demanding for others to focus on the details of the trees in each (otherwise they're "painting everything as the same shade of grey") is just a way to try to avoid that people look at the forest as a whole.
That said, you're right. The details are different and I didn't address that in my original post were I only talked about the main policy direction on these domains.
The broad policy direction on this subject is the same and the outcomes have been very similar and over time progressed in the same direction during the time in power of both parties, but things worsened in different domains at different speeds with different parties in power.
This is not even what many Americans call "the ratchet effect", it's actually worse because in this case it's not one pushing in a certain direction and the other refusing to revert it, it's actually both pushing in the same direction, with just some difference in details here and there which didn't add up to much difference in outcomes.
So yeah, my point stands that in this domain both US parties are shit and my second point also stands that you're trying to move the conversation away from criticizing parties for doing this shit by claiming that subtle differences in each party's shit are more important that the overall shitty nature of their actions in this.
What are the laws about search warrants around home cameras and the 5th amendment?
I’ve thought about setting up old smart phone based IP cameras around my house facing out windows. But decided that if it comes down to arresting people for anti regime speech, that having cameras with background audio of private conversations wasn’t a good idea.
I’m not sure it matters if it’s legal or not anymore these days.
Still, they can legally demand any recordings from you if they reasonably can know that such recordings exist. Generally they will need a warrant or they may subpoena you for the evidence that they know you have. You can even be arrested for erasing your own footage as destruction of evidence.
Obligatory statement that I am not a lawyer and this isn’t legal advice.
They can only get it with a search warrant. If everything is encrypted with a sufficiently strong password, I think the court precedent is that they can't compel you to reveal the password.
To get a warrant, they need to convince a judge that it's necessary to prove guilt in a specific crime, which means they need at least reasonable suspicion before even asking for the footage.
Yeah, really my question should have been about encrypted footage and my 5th amendment to protecting the password to the footage.
Hopefully no one needs to test this to find out.
The question for smartphones has been tried in court IIRC. Basically, police can compel you to unlock your phone with biometrics, but cannot compel you to unlock it if it's a password, and the difference is your fingerprint is something you have, whereas a password is something you know. Your fingerprint is subject to the fourth amendment and your password is subject to the fifth.
So when it comes to video footage, the password is protected, so they'd need to break the encryption or the password, they couldn't compel you to reveal it.
Go Colorado Sun! Proud sponsor for many years!
Reading the article, I am very confused. It appears that they simply decided a random person was the culprit because she was recorded as driving through town during the time period of the package theft, and that's all they had?
Worth noting that Ring has announced a partnership with Flock.
cnet.com/home/security/amazons…
So if you're in the Ring ecosystem, maybe time to re-consider.
Amazon's Ring Cameras Push Deeper Into Police and Government Surveillance
Ring has partnered with Flock Safety, making it easier for law enforcement to reach out to Ring doorbell and security camera owners to request footage.Omar Gallaga (CNET)
No wonder Stephen Miller is so against citizens wearing masks.
Edward Snowden did try to warn us over a decade ago.
I strongly encourage everyone interested in this topic (and you should be!) to read the article because this shit runs deep and they see absolutely no problem approaching the law in this fashion. Absolutely disgusting erosion of liberty and privacy, though it's not the least bit surprising. Here's an excerpt i found particularly chilling--this cop is fully convinced (or acting as if he were) about the validity of this minimal-effort investigation they apparently were ready to arrest someone over. Note that weeks later it was fully disproven and ended with a terse email acknowledging that she provided enough proof to absolve herself as the suspect. No accountability for their mistake, just: "you can go now"
“You know we have cameras in that town. You can’t get a breath of fresh air in or out of that place without us knowing,” Milliman said to Elser, according to Ring doorbell footage of the Sept. 27 encounter viewed by The Colorado Sun.“Just as an example, you’ve driven there about 20 times in the last month,” he added.
Along with the Flock footage, the sergeant told Elser he also had a video from the theft victim that allegedly showed Elser ringing the doorbell before grabbing a package and running away.
My favorite part
“I guess this is a shock to you, but I am telling you, this is a lock. One hundred percent. No doubt,” Milliman said.
😳
But Elser, a financial advisor, told the sergeant she had no idea what he was talking about. She asked several times to watch the video that Milliman insisted proved her guilt, but he refused to show her. And when Elser offered up footage from her Rivian’s onboard cameras to prove her innocence, Milliman said she could bring it to court.“It doesn’t matter. I’ll be giving this all to you. If you are going to deny it to me, I am not going to help you with any courtesy,” Milliman said.
“It’s kind of funny because we have cameras on our truck, so we could show you exactly where we were,” Elser said.
We are really fucked here. No accountability on their end, while foisting 200% accountability on ours.
like this
warriorbob likes this.
Yeah, been like this for quite a while. They can drag you for a while, lose their case, shrug it off, and continue as normal.
Meanwhile, you lost your job after your arrest, maybe even were denied bail and had to stay ~2 years in jail waiting for trial, and spent $100k on legal expenses. Winning at trial gives you no restitution for those massive losses. You're expected to also shrug it off and continue life.
Sometimes lawyers do preliminary motions like to suppress unconstitutional search warrants or change of venue and stuff. If it's complex, it can take a while, and defense cannot request speedy trial if they're filing things, but you also don't necessarily want to forgo filing useful things.
Also, if they violate the constitutional right to a speedy trial, you can file a habeas corpus or something and, even if you win, there's still no consequence except them shrugging and saying oops.
This reminds me of how police abuse any new tool they're given.
Like how while trained dogs can actually sniff out drugs, when they're given to police, they get retrained to simply alert whenever the police want them to, and essentially become a flimsy reason to let police violate your rights and search anybody they want to.
And the police suffer zero repercussions for their actions. If they don't find drugs, there's nobody who's going to take them to court and force them to retrain their dogs or to disallow drug dogs from being used as reasonable suspicion.
We are really fucked here. No accountability on their end, while foisting 200% accountability on ours.
Is there some reason victims can't just sue flock into oblivion?
Absolutely disgusting erosion of liberty and privacy, though it’s not the least bit surprising.
Legally, it's not an erosion.
Public spaces aren't private, and it was a charge that hadn't yet reached (probably costly) trial.
It's the same level of erosion as before when they lacked this level of public surveillance.
this cop is fully convinced (or acting as if he were) about the validity of this minimal-effort investigation they apparently were ready to arrest someone over.
That's standard procedure for police in the US: overconfidence & pressure of any kind (eg, lies) to extract a confession no matter if false or the evidence doesn't support it.
Their approach seeks conviction (no matter what) rather than truth.
They're twats.
No accountability on their end
Their unaccountability is standard.
Welcome to US law enforcement.
They were just as bad before.
:::spoiler Apparently, policing can be better.
UK policing was similar to the US until legal reforms (due to high profile cases of coerced confessions) led them to develop investigative interviewing, which seeks to gather evidence (free from biases & contamination) rather than confessions.
Much of the scientific base of investigative interviewing stems from social psychology and cognitive psychology, including studies of human memory. The method aims at mitigating the effects of inherent human fallacies and cognitive biases such as suggestibility, confirmation bias, priming and false memories. In order to conduct a successful interview the interviewer needs to be able to (1) create good rapport with the interviewee, (2) describe the purpose of the interview, (3) ask open-ended questions, and (4) be willing to explore alternative hypotheses. Before any probing questions are asked, the interviewees are encouraged to give their free, uninterrupted account.
When mandatory recordings revealed officers were unskilled interviewers (eg, assumed guilt of interviewee) missing & ignoring evidence due to their biases, and therefore needing training
they devised a program called PEACE with the help of psychologists. The week-long course, which also covered interviewing witnesses, was undertaken by every operational officer in the country. In the UK, unlike the USA, there is a high degree of cooperation and standardization between all forces. The training was a massive commitment, but it has helped avoid miscarriages, and it delivers better justice. Research studies and practical evaluations have also consistently shown higher skill levels and more objective approaches by officers. It is now accepted that not all officers will make good interviewers. PEACE has developed into several tiers of training linked to an officer’s field of work and identified potential.
Moreover, they refrain from lying.
The law does not allow lying to suspects, under any circumstances. Officers are trained to concentrate on probing a suspect’s account, seeking to confirm or negate by comparison with other known information. When the suspect knows that I can’t lie—my job is on the line if I do—I get more information.
:::
where can I find one of these? asking for a friend.
will a blueray from a burner work, or do I need the get a green laser capable of taking out a pilots vision for this?
Seems like some kind of oily fog/spray could obscure things until someone took the time to physically clear it. More temporary, but perhaps easier to accomplish?
Edit: And this pisses me off that I have to think about such things.
The root of the issue is allowing officer to lie in order to deprive people of thier rights.
He knew he had nothing, he was just trying to get a confession by saying it was a 100% lock. The cameras wouldn't matter as much if lieing like that was illegal.
like this
warriorbob likes this.
The fact that police officers can lie but people can't shows you the terrible power imbalance in our law enforcement.
Important for those who don't know: police can legally lie to you. Happens all the time when they're trying to get a confession. In a discussion, they'll be like "we have your fingerprints matched and we have video of you, so it's better if you're just honest with us." But they often don't have anything which is why they're desperate for a confession.
Weird to me that people are taking issue with the cameras more than the police work.
The problem here is charges being made with weak evidence and officers legally allowed to lie. I had a similar experience, but she was smarter than me. I was 22 and naive, thinking I didn't need to prove my innocence because they have to prove my guilt in court (logically seemed impossible when I wasn't guilty). The presumption of innocence is a lie. And juries and judges don't operate with pure logic and reason. I had to learn the hard way, losing many years of my life.
like this
demi_demi_demi likes this.
And that's why you DON'T TALK TO THE POLICE.
If you are detained, do not talk at all, even if you're nervous, even if you think you're being helpful. Do not volunteer anything. If you are arrested, you say exactly this and nothing else: "I invoke my right to remain silent, and I invoke my right to an attorney." Repeat that exact phrase AND NOTHING ELSE until you have your attorney present.
I got pulled over the other day. The reason given was a lane change violation (which was bullshit pretense, it was right outside a very rural, but very busy, bar so this was likely actually entrapment, tho I was for sure under the legal limit - I was there to check out line dancing because I’ve never seen it before, and only had one beer in the hour I was there).
I also had a very expired registration (haven’t driven much, and didn’t realize I forgot to renew it).
But I got let off everything with a warning..? I spent days trying to figure it out because it should have been a ticket.. he didn’t even seem interested in waiting for me to dig out my insurance info (which I had, just had to get it out of my wallet).
But I have a dash cam.. and it records sound. It would have proven I didn’t violate anything, and he was recorded saying why I was pulled over so no way to flub it and say it was actually the registration all along, and thus the pretense for pulling me over in the first place was void. I’m pretty sure that’s the only reason I got off with a list of warnings rather than tickets.
This is exactly the tactic the officer was employing here (for a sub $25 theft), not showing the accused the evidence so they don't know what the police might or might not know.
At some point in the process, there is "discovery" where both sides share their evidence before trial to avoid going to trial for stupid stuff (like this.) But you usually have to engage thousands of dollars of legal services before discovery is available, again over a sub $25 theft allegation.
The officer sweating her for driving through his town on the day somebody porch pirated somebody else is really ridiculous.
that's exactly what I got out of this whole situation.
guilty until proven innocent.
That's how they're running it, and there are a whole lot of people who would prefer it to run that way in the future.
What should be happening is: when falsely accused and exonerated in court, you get a judgement against the LEA for treble damages for your costs to rebut their false claims.
False claims are going to happen, but if they're costing the police thousands of dollars per instance, that should slow them down. I'm more than happy to pay increased taxes to put that deterrent on the agencies.
Yes, that's the point. Their glass ball and Tarot layout say you're guilty, so now you have to prove your innocence. And to prove your innocence you have to collect all the data on yourself.
BTW, this is far more subtle than it seems, collecting and giving to someone all the info on yourself all the time is nonsense, but collecting it and having just in case for such situations might seem normal for many honest people. Except in fact these are the same, you don't have tools to collect it all without giving it to someone predictable. So this whole big tech and surveillance con abuses good faith participation in the society. And encourages everyone becoming a cheater.
The police and other such people know that these are bullshit machines, but use them to cheat with impunity. Sometimes to charge a clearly innocent person, because they have an excuse - the computer did it. And the rest of us are incentivized to cheat to get better ratings for loans and worse ratings for scammers, and better danger rating so that police wouldn't just use as a scapegoat to close a case like this, instead choosing someone less dangerous.
Wait till witchcraft becomes a crime again. Nobody would believe in it, of course, but it'd be an easy win for everyone except the convict.
I don't care if Soviet caricatures ("Neznaika on the Moon" specifically) were wrong back then, they are correct now. I mean, yeah, they are correct everywhere now, but still.
So because she is better off financially and is not worried about google tracking, she had all the cameras, GPS tracking, and everything set up to prove her innocence.
I decline all of that stuff and i would have a MUCH tougher time proving my innocence when wrongly accused like she was.
This is just another step towards fascism where police are charging people for crimes they never committed, based on AI and computers screwing up.
That's intentional. Someone just makes shit up, using a magic machine, so that their responsibility were in doubt for other similar irresponsible people with ability to fuck up others' lives.
There should be a responsible policeman for every such decision, going to jail for at least as much time as she would were she convicted, when the decision is wrong.
She feared the impact a theft charge, though small, would have on her financial career.
Wild that a false accusation, after being proven as false at the court of law, can still impact one's career.
Yeah, that's something that absolutely has to change. I don't care if "career criminals get out of charges all the time". A false charge should not follow you for the rest of your life.
Then again, I also believe that if you serve your time in prison and are released, you should not have a publicly searchable record that can be used to deny you opportunities. So take my opinion as you will
"what about repeat violent offenders."
This is conservative paranoia propaganda at work. People who are violent offenders become repeat violent offenders because of the system that we have in place not in spite of it. And the percentage of violent offenders in our prison system is severely out of proportion to those of the nonviolent variety who make up the bulk of our inmate population.
Agreed, and prison should be for rehabilitation.
Perhaps prisoners could be released in one of two states: completed time or rehabilitated. The latter carries a much lower chance of recidivism. Maybe the first iffense could be hidden regardless, and expunged entirely after some period of time (10 years?), whereas on the second offense, both are searchable.
IDK, but I do believe in forgiveness.
Wow. In ex-USSR past convictions are a problem, but when you were cleared of charges - that really is wild. I mean, OK, the rate of convictions is not exactly normal in ex-USSR too.
I mean by this comparison that people here usually think we have it worse with the conviction record.
Why can't they see the outcome?
AI is built on a reward system. Its sole reason for existence is to complete its task and get the reward points. It will create false information to do this. One AI that a lawyer "accidentally" used in court actually created its own 4-5 page court cases to use as citations to justify the case it was working on.
AI is a novelty and should NOT be in charge of any decision making or be admissible as evidence in any way.
Yep. Dogs have been used to manufacture probably cause for decades.
Only once have they ever been scientifically tested, and they failed.. and shockingly, cops refused to participate in any future testings.
As the owner of a German shepherd who just REALLY wants to make friends and play with everyone she meets…it’s depressing how many people see a big cop dog and immediately walk away when she barks.
She wants to chase birds and lick your face to show affection, chasing and hurting people is taught just like racism in humans.
Dog owner here. I don’t know if I buy the whole “don’t judge by breed” thing. Sure, training can become the dominant force, but dogs are literally wolves that were selectively bred based on temperament. And how would genes decide so much about a dog but not its temperament?
Anyway still sucks. I’d want to hang out with your dog. But I respect where people are coming from.
Also cool username. Although makes me think of some weird HGTTG marital arrangements.
Oh, they should, but similarly to "AI" as a tool, with the whole responsibility for the tool being on the person using it.
Similar to screwdrivers, pencils and guns.
AI also recently decided a bag of chips that a black kid had was a gun, and summoned a horde of cops on him.
an accident I doubt AI would make with a white kid, because AI gets all sorts of inherit biases from the data its fed.. and whats more biased in law inforcement than how black people are treated vs white people.
Did they? That's strange considering how prez has let them off the leash with Ai regulation. In that case I'm curious if the data centers that are being built now will be the only ones.
Edit: Oh, you meant in CO. That's news to me too
Everyone freaking out has forgotten: Do not talk to the police. Guilt is determined in court and anything you say, drumroll please, can be used against you. You will not talk your way out of getting arrested, shut the fuck up, and sort it out in court. The only person there to help you is your lawyer, the police are not there to help you.
She feared the impact a theft charge, though small, would have on her financial career.
Why is this info public, what happened to innocent til proven guilty?
innocent til proven guilty
That only works, inside the court.
Outside, if you come in the view of an officer, you are guilty.
I have had to do something similar recently, because some chap with "senior citizen" status randomly blamed me for something.
Justice Department puts 2 prosecutors on leave after they signed court docs that described "mob of rioters" on Jan. 6
The Justice Department placed two D.C.-based federal prosecutors on leave after they filed court papers calling the Jan. 6 Capitol siege a "riot" perpetrated by a "mob," three sources familiar with the matter told CBS News on Wednesday.
The papers were submitted Tuesday in the case of Taylor Taranto, who was pardoned by Trump on Capitol riot charges earlier this year but was later convicted of livestreaming a bomb threat. He was arrested in 2023 while livestreaming himself driving around former President Barack Obama's D.C. neighborhood while armed, according to prosecutors.
The filing — which asked a judge to sentence Taranto to 27 months in prison at a hearing Thursday — mentioned Taranto's Jan. 6 charges and briefly described the events of that day, writing that "thousands of people comprising a mob of rioters attacked the U.S. Capitol."
That unsparing description of the Capitol riot was notable, as Mr. Trump has called Jan. 6 a "day of love" and referred to the rioters as "hostages."
Justice Department puts 2 prosecutors on leave after they signed court docs that described "mob of rioters" on Jan. 6
The Justice Department placed two federal prosecutors on leave after they filed court papers calling the Jan. 6 Capitol siege a "riot" perpetrated by a "mob," three sources told CBS News.Scott MacFarlane (CBS News)
like this
adhocfungus e essell like this.
Los Lobos - Gates Of Gold (2015)
Questo disco arriva dopo che le ultime prove discografiche in studio erano diventate un poco appannate, avevano perso smalto (“The Town and The City” una spanna sopra l’ ultimo “Tin Can Trust” di cinque anni orsono, tuttavia entrambe sono prove meno convincenti di un glorioso passato)... Leggi e ascolta...
Los Lobos - Gates Of Gold (2015)
Questo disco arriva dopo che le ultime prove discografiche in studio erano diventate un poco appannate, avevano perso smalto (“The Town and The City” una spanna sopra l’ ultimo “Tin Can Trust” di cinque anni orsono, tuttavia entrambe sono prove meno convincenti di un glorioso passato). Però David Hidalgo e Louie Pérez sono tornati in gran forma e c’è una grande varietà nei suoni, con brani che si presentano in una veste squisitamente latina, oppure troviamo blues urbani, sonorità black, poi ci sono riferimenti ai Grateful Dead ed al loro capolavoro Kiko e riscontriamo la presenza di alcune grandi canzoni come ad esempio “Magdalena” e “When We Were Free”. Azzarderei col dire che è il disco più convincente dai tempi di “Good Morning Aztlán” (2002) e qualitativamente siamo ai livelli di Kiko... artesuono.blogspot.com/2015/10…
Ascolta il disco: album.link/s/7oM8JtjRTcDm4F9I3…
Home – Identità DigitaleSono su: Mastodon.uno - Pixelfed - Feddit
Los Lobos - Gates Of Gold (2015)
di Massimo Orsi Questo disco arriva dopo che le ultime prove discografiche in studio erano diventate un poco appannate, avevano perso s...Silvano Bottaro (Blogger)
RRF Rassegna stampa 30 10 25. Corte dei conti boccia piano ponte sullo stretto. Ira del governo. Opposizioni "Parole gravi"
like this
BrikoX e adhocfungus like this.
reshared this
Feddit Un'istanza italiana Lemmy reshared this.
like this
adhocfungus e Maeve like this.
reshared this
Feddit Un'istanza italiana Lemmy e Technology reshared this.
The reporting goes pretty hard:
‘No restrictions’ and a secret ‘wink’: Inside Israel’s deal with Google, Amazon
To secure the lucrative Project Nimbus contract, the tech giants agreed to disregard their own terms of service and sidestep legal orders by tipping Israel off if a foreign court demands its data, a joint investigation reveals.
May this be a reminder to degoogle your phone and boycott amazon. It's not hard, and there are thousands of people on here eager to help. 😀
like this
Maeve likes this.
Microsoft seemingly just revealed that OpenAI lost $11.5B last quarter
Microsoft seemingly just revealed that OpenAI lost $11.5B last quarter
: Satya has also delivered Sam most of the cash he promisedMatt Rosoff (The Register)
Microsoft seemingly just revealed that OpenAI lost $11.5B last quarter
Microsoft seemingly just revealed that OpenAI lost $11.5B last quarter
: Satya has also delivered Sam most of the cash he promisedMatt Rosoff (The Register)
like this
adhocfungus, Scrollone, magic_lobster_party, IAmLamp, avernus, Oofnik, Maeve, MyTurtleSwimsUpsideDown e demi_demi_demi like this.
reshared this
Technology reshared this.
Investment is done really to train models for ever more miniscule gains. I feel like the current choices are enough to satisfy who is interested in such services, and what really is lacking is now more hardware dedicated to single user sessions to improve quality of output with the current models.
But I really want to see more development on offline services, as right now it is really done only by hobbyists and only occasionally large companies with a little dripfeed (Facebook Llama, original Deepseek model [latter being pretty much useless as no one has the hardware to run it]).
I remember seeing the Samsung Galaxy Fold 7 ("the first AI phone", unironic cit.) presentation and listening to them talking about all the AI features instead of the real phone capabilities. "All of this is offline, right? A powerful smartphone... makes sense to have local models for tasks." but it later became abundantly clear it was just repackaged always-online Gemini for the entire presentation on $2000 of hardware.
like this
Maeve likes this.
They're investing this much because they honestly seem to think they're on the cusp of super intelligent AGI. They're not, but they really seem to think they are, and that seems to justify these insane investments.
But all they're really doing is the same thing as before but even bigger. It's not going to work. It's only going to make things even more expensive.
I use Copilot and Claude at work, and while it's really impressive at what it can do, it's also really stupid and requires a lot of hand holding. It's not on the brink of AGI super intelligence. Not even close. Maybe we'll get there some day, but not before all these companies are bankrupt.
like this
TVA likes this.
Comparing the coming crash to the dot com crash is like comparing a rough landing to the various crashes on Sept 11th, 2001.
The dot com crash was mostly isolated in high tech. Because it was lead by the Japanese economy starting to fail, and followed by the Sept 11th attacks, the various combined crashes resulted in the S&P 500 falling by about 50% from its peak to the bottom, but it was already back up to the peak value in 2007, then the global financial crisis hit.
This bubble is much bigger. Some analysts say the AI bubble is 17x the size of the Dot Com bubble, and 4x the size of the 2007/08 real estate bubble. AI stocks were 40% of all US GDP growth in 2025, and 80% of all growth in US stocks.
Nvidia's stock price has gone up 1700% in just 2 years. OpenAI is planning to go public on a valuation of $1 trillion despite losing vast amounts of money. Just 7 US tech companies make up 36% of the entire US stock market, and they're all heavily betting on AI.
At least when the dot com bubble popped, it left some useful things behind, like huge amounts of dark fibre. But, the AI processors are so specialized they can't be used for much of anything else. They also wear out, sometimes within months. The datacenter buildings themselves can maybe be repurposed to being general purpose datacenters, but, a lot of the contents will have to be thrown out.
The Entire Economy Now Depends on the AI Industry Not Fumbling
Without AI spending, the US economy would lose one of the last industries actually bringing in any healthy revenue.Joe Wilkins (Futurism)
The problem is there is little continuous cash flow for on prem personal services. Look at Samsung's home automation, its nearly all online features and when the internet is out you are SOL.
To have your own Github Copilot in a device the size and power usage of a Raspberry Pi would be amazing. But then they won't get subscriptions.
more development on offline services
There is absolutely massive development on open weight models that can be used offline/privately. Minimax M2, most recent one, has comparable benchmark scores to the private US megatech models at 1/12th the cost, and at higher token throughput. Qwen, GLM, deepseek have comparable models to M2, and have smaller models more easily used on very modest hardware.
Closed megatech datacenter AI strategy is partnership with US government/military for oppressive control of humanity. Spending 12x more per token while empowering big tech/US empire to steal from and oppress you is not worth a small fraction in benchmark/quality improvement.
what really is lacking is now more hardware dedicated to single user sessions to improve quality of output with the current models
That is the exact opposite of my opinion. They're throwing tons of computing at the current models. It has produced little improvement. The vast majority of investment is in compute hardware, rather than R&D. They need more R&D to improve the underlying models. More hardware isn't going to get the significant gains we need
I wonder how long it will be before investments start getting pulled back because of a lack of ROI.
Just wait for the next hot thing to come out
One of our biggest bookstores contracted with a local artist for some merch. That artist used AI with predictable results. Now everyone involved is getting raked over the coals for it.
No surprise, they just announced a 4th round of layoffs too. 😟
lithub.com/everything-you-need…
koin.com/news/portland/powells…
Everything you need to know about the Powell’s AI slop snafu—and what we can all learn from it.
Another day, another duel with AI slop. Unfortunately a recent deep fake has come in the guise of a friend. Last Friday, Powell’s Books of Portland, Oregon—one of the country’s best lov…Literary Hub
AI is funded solely by sunk cost fallacy at this point.
and the us economy an gdp relies solely on ai make of that what you will.
like this
TVA likes this.
That's the only reason I don't think it will pop in the next 6 months or so. Even Biden or Obama would have stepped in to try to prevent the economy from crashing. But, there's the Trump factor. First of all, some of his biggest backers are from the AI "industry". His VP is tied to Peter Thiel, his biggest donors are Crypto and AI bros. The vast majority of his own personal money is tied up in the current Crypto bubble. In addition, he's obviously so easily bribed. Even if he he wasn't interested in intervening otherwise, he could easily be bribed to intervene.
Because of Trump, and the fact that the house, senate and judiciary are all Trump lackeys, I think the bubble will survive until at least the 2026 midterms. If the Democrats take back control of the House and Senate they could take control over spending from Trump, which might mean the bubble is allowed to pop. But, I wouldn't be surprised to see Trump hand over literal trillions in taxpayer dollars to keep the bubble inflated.
this is not a bad analogy, but you are off by orders of magnitude
more importantly, both Uber and Amazon always had a path to profitability (Amazon specifically was already making tons of money on AWS long before the store front made money). AI has already been shown to not have a path to profitability; whatever little value companies around the world have been able to extract, cannot pay the cost of producing it.
think of it this way:
You produce a little car that can drive 2 people and some bags around, it costs you $1000 to make and you sell it for $3000 which a ton of people can afford... you have a path to profitability
I enter the market with a car that can carry 20 people, plus full on luggage for all and it moves twice as fast... but, in practice, I can only really move 3 people and often take them the wrong way, also the luggage was a complete lie and I can only allow passengers with their purses... also my car cost $50,000 to make so I would have to sell it for $70,000 and nobody would pay that when they could get 20 of your cars for less... also also, I promised the people making some parts of my car that would invest 7 kajillion on their companies somehow.
Which company would succeed? yours or mine?
Why do you think AI is pushed so hard?
Everyone is aware this has to be useful. Too much money.
Still the powers that be will do everything to avoid a hard crash, which would be so much earned.
I wouldn't have a problem if they were actually investing the money in something useful like R&D
Nearly all the investment is in data centers. Their approach for the past 2 years seems to be just throwing more hardware at existing approaches, which is a really great way to burn an absurd amount of money for little to nothing in return
It’s very corporate, isn’t it? “Just keep scaling what we have.”
That being said, a lot of innovation is happening, but goes unused. It’s incredible how my promising papers come out, and get completely passed over by Big Tech AI, like nothing matters unless it’s developed in house.
The Chinese firms are picking up some research in bigger models, at least, but are kinda falling into local maxima too.
But that's what they wanted anyway, isn't it?
Burning shitloads of money.
Waiting until they can later, finally, rule the world.
Problem is they are competing with cheap web services like deepseek and local free models. Those alternatives are gonna become more popular when chatgpt starts charging.
They are spending like crazy in the hope for some inovation that will give them an advantage that others can't copy for cheap. That is a very difficult thing to accomplish. I bet they will fail. That money ain't coming back.
like this
Maeve likes this.
If you owe the bank $100, that’s your problem; if you owe the bank $100 million, that’s the bank’s problem.
like this
Maeve likes this.
You need to be as precise as your resolution, otherwise the precision is meaningless. I guess you could argue that your resolution is units of half-billion (since some things are measured like that), but the initial value of 0.1B, and your use of 0.5 rather than 'half' suggests a resolution of 0.1B.
This is different to the aphorism 'The difference between a million and a billion is about a billion', both because of the difference in scale, and the quoted resolution.
So I wondered a bit how much it actually affects the economy.
"S&P 500" companies' market cap is about 57 trillion dollars with a P/E ratio of about 30. So openai by itself is dragging down the total s&p 500 earnings by only about 0.5%. The bigger problem is that there are multiple companies like openAI, and a large chunk of the entire economy's valuation is tied to the promise that all the AI companies will somehow become profitable sometime soon.
like this
Maeve likes this.
The Hater's Guide To The AI Bubble
Hey! Before we go any further — if you want to support my work, please sign up for the premium version of Where’s Your Ed At, it’s a $7-a-month (or $70-a-year) paid product where every week you get a premium newsletter, all while supporting my free w…Edward Zitron (Ed Zitron's Where's Your Ed At)
I was referring to the general concept behind the quote.
I originally want to post the OG (apocryphal?) variant:
Owe Your Banker £1,000 and You Are at His Mercy; Owe Him £1 Million and the Position Is Reversed
But it sounds rather quaint these days.
In contrast to the housing bubble, where a lot of the value was in overpriced houses sold to individuals, this overpricing is almost entirely in tech stocks, and tech stocks are almost entirely owned by by the wealthiest 10%, even 1%. The tech billionaires have limited ability to divest themselves of their own overpriced companies and absolutely will lose money.
None of them are going bankrupt, they'll all be just fine when the market recovers in a few years, because that's the nature of capitalism. A bunch of peons, who convinced themselves that the bubble-value of their 401k meant it was safe to retire, will suffer, will have to go back to work - if you're not an oligarch, losing money is painful.
The bubble was accelerated by people assuming prices would always rise, often their own greed and flipping or becoming landlords,, and this was only possible due to the financial engineering that made lending to these folks posivle, but all of that droce the value of the underlying assets down significantly.
You claim there wasn't a bubble...look at home price values in the sunbelt in 05 vs 2009.
TF are you talking about there wasn't a bubble
this was only possible due to the financial engineering that made lending to these folks posivle
There was no "financial engineering". It was just fraud.
look at home price values in the sunbelt in 05 vs 2009. TF are you talking about there wasn't a bubble
"Bubble" does not mean "prices go up and then back down again".
In contrast to the housing bubble, where a lot of the value was in overpriced houses sold to individuals,
was?
since the housing bubble collapsed
did it? it does not seem so. where I live to buy a house in good condition people need to take out loans that the bank may not even allow, but if it does they'll pay it for decades. even empty plots are still very expensive. more and more people live in a rented place even though they don't want to move, because their house is taken away.
Yeah all the people praying for a crash are praying for nobody to have retirement funds.
You can easily tell who's actually employed in this thread because anyone with a 401k is going to get dicked down while the 0.1% get a bailout.
Remember when OpenAI launched Dall-E 2? You got a few tokens for free images and then had to pay for it. Presumably that was at least some reflection on the cost of producing the images.
Now you can create video for free and consumer expectations that generative AI should be super cheap have been set. That genie is not going to go easily back into the bottle.
like this
TVA e demi_demi_demi like this.
like this
TVA e demi_demi_demi like this.
like this
Maeve likes this.
like this
Maeve, TVA e demi_demi_demi like this.
Feels like the entire AI industry is built on "don't worry, growth will save us", but at some point someone has to pay the electricity bill...
like this
Maeve, TVA e demi_demi_demi like this.
like this
TVA likes this.
Im watching all this and im thinking you guys are being convinced to not buy these stocks.
I just keep buying. Because I know how this works. Seen it many times now. Media says one thing, market goes another way.
Do you guys remember when Tesla was down extreamly much because the media and most people on Lemmy were convinced that musks nazi salute would mean the company is done? It was only like six months ago, you cant have missed it.
Now go look at the stock. Buy tech stocks. Dont do what the media is telling you. They are not there to help you.
You have to be smart, and being smart is not following the public opinion about things, because its being shaped by the owner class. If the media is telling you to sell, you probably should be buying. And when its telling you to buy, be very careful because the media owners want to sell.
You have to view it like you are inside the matrix. In a way, we are.
Of course its not enough. Stock market cares about money, not morals. They know negative emotions are temporary and doesnt last.
I see more white Teslas than ever on the streets. They are the most popular company leasing vehicle appearently. A lot of things play into what cars people pick. Most people probably forgot about the nazi salute in a few weeks, or got a good deal on a Tesla and bought it anyway.
I dont think this is a bubble and it wont burst. I think its the beginning of a switch to a robotic society, with robots, Ai and implants, all requiring more energy than we have right now. The massive investments in power may look like its going to be used for training models, but it will be used for very different things we dont even have right now.
Any future high tech society will need tons of power. The billionaries are several steps ahead of what the media is talking about. As always.
Companies and governments are gonna buy robots, made by other robots. Typical scenarios include military, manifacturing, surveillance and so on.
Today companies pay humans for those things. It will change as soon as robots can do a task.
Virtual Reality? Not in the way we have it now with huge expensive glasses and special software. Thats just the first experimental phase of the technology.
I dont think it has a purpose yet. Maybe in the future, it will. Some killer app where VR is essential and everybody wants that app.
Probably something related to porn. Wouldnt be surprised.
I was making a parallel to another wildly over-hyped technology that has had multiple opportunities to make it when it's clearly only suitable for niche usecases.
LLMs and "AI" are not useless but the notion that they'll lead to something significantly more advanced is fundamentally misunderstanding the nature of the technology.
Y’know, I’ve been hearing that drumbeat for well over a decade. News doesn’t sell if it isn’t bad. Eventually the Boys Crying Wolf will be right, but it certainly won’t be because of their prognosticating accuracy. The market has long stopped making sense.
You cannot know what the market will do, and if you can figure it out, it’s already too late. If you want to assume the parent comment will actually lose money because they didn’t put stops in place or whatever, fine. My comment is still correct: If you sell above purchase price + fees and gains taxes, you don’t “lose” money. The rest is just making stuff up.
What's investment got to do with the article (or even the comments or people's sentiment on the Threadiverse)?
There are a wide variety of investment strategies depending on your situation and other factors. I don't see how these are related to tech news discussions.
I am not saying your are wrong or right, just an observation.
Perhaps, but how do you know that this is an actual trend?
An argument can be made that there are more people with your sentiment or perhaps the news about a bubble will attract short terms investors (trying to cash in as the wave is rising).
Generally speaking, discussion on Threadi, even ones that mention bubbles in context of AI/Nvidia, don't mention investment strategies.
I dont know if its an actual trend. But it would make sense that most people buy and sell based on what the media is telling them. I believe that most people trust what the media writes. They are not sceptical and cynical like me. 😀
I dont have an investment strategy except buy and hold.
Of course I am influenced by various things, we all are. I am not implying otherwise.
On a personal level, I don't see the connection between posting on a forum about articles on the possibility of an AI bubble and investment decisions.
I am not necessarily disagreeing with you. I think people are a bit more complicated than that and circumstances can be very different.
The key is getting out at the right time, and that is weighed massively against small investors. The big investors and institions control the market and can move quickly while small investors cannot.
Tesla is not doing well - look at its falling sales. It's a risky stock to hold. The AI companies are also highly risky stocks to hold.
That doesn't mean don't hold them - all anyone is saying really is that these are high risk investments, and at some point they are going to probably crash because it's a bubble.
That doesn't necessarily mean "don't invest". It does certainly mean be prepared to get out fast and also only use money you can afford to lose when investing with such high risk stocks.
like this
TVA likes this.
If you look at Tesla chart, or any other tech stock chart, you can see that nobody would have lost money if they didnt sell. They are all going up long term. Tesla is up 100% in three years, 240% in five years. Everybody who sold during that time made the wrong choice, unless they invested in something that went up more.
Just dont sell, ever, until you retire. This is super hard to do but its what the charts show is actually best.
Nothing will help you when you are shot dead in a camp. You live your life until that happens, if that happens.
But yeah, I understand your principles. The problem is that if only bad people have money, only bad people have power as well.
I missed both gold and defence... Now I dont know, what is the next thing to invest in? What do you think?
I know power is something that will be massively needed.
like this
TVA likes this.
like this
TVA likes this.
And even though NVIDIA is better place as they do produce something, but the something in play has little value out of the AI bubble.
NVIDIA could be left holding the bag on a super increased capacity to produce something that nobody wants anymore (or at least nowhere near at the levels we have now) so they are still very much exposed.
but the something in play has little value out of the AI bubble.
You're delusional if you think GPUs are of little value. LLMs and fancy image generation are a bubble.
The gargantuan computational cost of running the machine learning processing that is now required for protein folding and molecular docking is not.
The gargantuan computational cost of running the machine learning processing that is now required for protein folding and molecular docking is not.
Sure but do you need the absolute gargantuan capacity that is being built right now for that? if so, for how long and at what cost?
The point is not that GPU per se are of little value... the point is that what would you do with 10,000 rocket ships if you only have 1000 projects that may be able to use them? and what can those projects actually pay? can they cover the cost of the 10,000 rockets you built?
me too, but the GPU used for AI are not the same as what we would use at home.
maybe the factories can produce both kinds and they would be cheaper, but it is speculation at this point
It’s literally the same chip designers, production facilities and software. Every product using <5nm silicon fabs compete for the same manufacturing capabilities (fab time at TSMC in Taiwan) and all Nvidia GPUs share lots of commonalities in their software stack.
The silicon fab producing the latest Blackwell AI chips is the same fab producing the latest consumer silicon for both AMD, Apple, Intel and Nvidia. (Let’s ignore the fabs making memory for now.) Internally at Nvidia, I assume they have shuffled lots and lots of internal resources over from the consumer oriented parts of the company to the B2B oriented parts, severely reducing consumer focus.
And then we have any intentional price inflation and market segmentation. Cheap consumer GPUs that are a bit too efficient at LLM inference will compete with Nvidias DC offerings. The amount of consumer grade silicon used for AI inference is already staggering, and Nvidia is actively holding back that market segment.
blogs.microsoft.com/blog/2025/…
The next chapter of the Microsoft–OpenAI partnership - The Official Microsoft Blog
Since 2019, Microsoft and OpenAI have shared a vision to advance artificial intelligence responsibly and make its benefits broadly accessible.Microsoft Corporate Blogs (The Official Microsoft Blog)
like this
TVA likes this.
IIRC, OpenAI lost $12b for all of 2024. CNBC reported that OpenAI restructuring this week, has MSFT not only with 27% equity stake, but a 20% royalty rate on revenue going forward which is certainly a nearly impossible hole for OpenAI to get out of.
OpenAI has published analytics on "suicide interactions" which proves that they mine their users data. Support for US military and Israel ensures that their mission is to destroy humanity, unless humanity pays it more to be more loyal to it. Everyone's "don't be evil" actually means "don't be evil unless fascism pays more"
The Israel/US empire needs OpenAI to build ever bigger/more comprehensive models that are even more expensive to use than ChatGPT/Sora's status as most expensive models. They need the analytics function to oppress population, and the empire is certain to side with OpenAI if it seeks revenue enhancement through theft of IP, including their users' IP. It is dangerous for anyone to use OpenAI services because theft and oppression by empire condoned symbiosis is by design. But the race for ever larger more expensive models means trashing the previous generation models quickly, which means no time for ROI from development.
OpenAI will need massive military/government contracts to support its $1T in investment promises. All of those are to opppress Americans/humanity. Meanwhile, government has just sponsored 9 independent supercomputer projects, plu8s is dedicated to the MechaHitler vision of reality, and so OpenAI must commit to unrestrained evil in order to get their fair share of the oppression mission, and survive. Expect US government contracts to develop models for it, but with OpenAI profitting from government and private surveillance use.
AI crash will be 11X bigger than subprime mortgage crash, also driven by mass stupidity.
The next movie will be titled: * The Big Shart*
This guys is jacked...jacked to the tits.
What's the deal with the "HPE" in some Register articles? It's apparently the Hewlett-Packard Enterprise logo, but articles about HPE don't appear to have that logo.
Is The Register affiliated with HPE now?
HPE puts all its chips in the agentic AI pot
HPE Discover 2025: Another OEM has decided we're now in the agentic AI ageBrandon Vigliarolo (The Register)
Billions in investment. Trillions in speculation. All on something that makes less money than Genshin Impact.
Fun times.
- YouTube
Profitez des vidéos et de la musique que vous aimez, mettez en ligne des contenus originaux, et partagez-les avec vos amis, vos proches et le monde entier.www.youtube.com
Oh honey, that hasn't been true since 2008.
The government will bail out companies that get too big to fail. So investors want to loan money to companies so that those companies become too big to fail, so that when those investors "collect on their debt with interest" the government pays them.
They funded Uber, which lost 33 billion dollars over the course of 7 years before ever turning a profit, but by driving taxi companies out of business and lobbying that public transit is unnecessary, they're an unmissable part of society, so investors will get their dues.
They funded Elon Musk, whose companies are the primary means of communication between politicians and the public, a replacing NASA as the US government's primary space launch provider for both civilian and military missions, and whose prestige got a bunch of governments to defund public transit to feed continued dependence on car companies. So investors will get their dues through military contracts and through being able to threaten politicians with a media blackout.
And so they fund AI, which they're trying to have replace so many essential functions that society can't run without it, and which muddies the waters of anonymous interaction to the point that people have no choice but to only rely on information that has been vetted by institutions - usually corporations like for-profit news.
The point of AI is not to make itself so desirable that people want to give AI companies money to have it in their life. The point of AI is to make people more dependent on AI and on other corporations that the AI company's owners own.
Well actually there is a long and rich history of companies that are able to operate at a loss using funds appropriated from sale of shares to investors, and this process continues so long as new investors keep buying in such that anybody selling out is covered by the new funds until enough people try to sell out that the price starts to plunge, although the collapse can be delayed by the company strategically buying back and occasionally splitting or reorganizing, meaning everyone gets their money back unless they sell too late.
You know.
A fucking Ponze Scheme.
same old
- YouTube
Profitez des vidéos et de la musique que vous aimez, mettez en ligne des contenus originaux, et partagez-les avec vos amis, vos proches et le monde entier.www.youtube.com
That's what happens when you don't listen to your customers.
technically according to NSPM-7 any FOSS is terroristic by nature because it's anticapitalist.
that means if you have contributed to FOSS at any time, you are a terrorist. technically.
I know this is not a real discussion 😁
But I don't think FOSS is inherently anticapitalist. It's just not late stage capitalism. There are plenty of commercial FOSS projects.
Sure you could compile them from source or download somones executable. But especially companies often want convenience, customer support and LTS versions.
Remote Controlled Crawler Robot - Built with ChatGPT Codex by a Non-Coder
- YouTube
Profitez des vidéos et de la musique que vous aimez, mettez en ligne des contenus originaux, et partagez-les avec vos amis, vos proches et le monde entier.www.youtube.com
Podcaster Jennifer Welch Denounces 'Dipsh*ts' Schumer and Jeffries for Not Riding Mamdani’s Progressive Wave
Podcaster Jennifer Welch Denounces 'Dipsh*ts' Schumer and Jeffries for Not Riding Mamdani’s Progressive Wave
The Democratic leaders, said the outspoken I've Had It podcast co-host, have refused to show up for the NYC mayoral candidate because they are "beholden to the same corporations that helped Donald Trump get elected."julia-conley (Common Dreams)
like this
adhocfungus, copymyjalopy e essell like this.
ICE deported a man who claims US citizenship after federal judge blocked his removal
Chanthila Souvannarath, 44, was born in a Thai refugee camp but has lived in the United States since he was an infant. He gained citizenship as a child when his father was naturalized, making him eligible for derivative citizenship under immigration law at the time, according to his attorneys.
...
On October 23, a federal judge blocked ICE from deporting him while he challenged his arrest and detention, but he was put on a plane for Laos the next day.
ICE deported an Alabama man who claims US citizenship. DHS says it wasn’t a mistake and don’t want him back
A federal judge had blocked ICE from deporting Chanthila Souvannarath, warning ‘inherent and obvious harm’ in deporting a citizenAlex Woodward (The Independent)
like this
copymyjalopy e adhocfungus like this.
RRF Sport. Basket serie B. Juve Caserta Luiss Roma 55 a 59
Viaggio fantastico tra pennelli di mosca e frammenti di polvere intagliati a misura - Il blog di Jacopo Ranieri
Viaggio fantastico tra pennelli di mosca e frammenti di polvere intagliati a misura - Il blog di Jacopo Ranieri
Una delle storie più frequentemente ripetute dall’artista britannico Willard Wigan nel corso delle sue interviste riguarda il processo formativo dei suoi anni scolastici ed il modo in cui un’esperienza negativa lo condusse, per via indiretta, alla la…Jacopo (Il blog di Jacopo Ranieri)
Bending Spoons to acquire AOL
Bending Spoons to acquire AOL | TechCrunch
Wednesday's news marks a new chapter for AOL, which was once one of the most recognized brands on the internet, known for its email service and "You've Got Mail" notification.Aisha Malik (TechCrunch)
like this
Mechanize, adhocfungus e RaoulDuke85 like this.
reshared this
Technology reshared this.
Can someone plausibly explain the Bending Spoons business model?
Big chunk of debt financing for decrepit brands with old tech. Other than strip-mining user data, I can't figure it out.
like this
fonix232 likes this.
like this
fistac0rpse e twelvety like this.
Doesn't mean they will keep the name for EU customers. The value there is in the IP and the infrastructure configuration.
Thus the term: Rebrand.
Europe Online Gets AT&T Dialtone
Computer service Europe Online and the Interchange Online Network – acquired by AT&T- recently announced an alliance to jointly conquer European cyberspace. AT&T’s Interchange will provide software for Europe Online, which hopes to challenge U.S.Variety Staff (Variety)
like this
dhhyfddehhfyy4673 e fonix232 like this.
This fucking company keeps buying things I use and ruining them. I can't believe I know their name
At least I don't use AOL
like this
fonix232 likes this.
Splice, impossibly expensive subscription even though I paid for a lifetime license (theft)
Focos, maybe they sold it, but they crippled the one thing I used and changed it to a subscription
Evernote, so long ago, I don't remember, but expensive
Komoot, soon to be too expensive and enshittified
It's just really weird I keep running into them with random apps I have used
Stephen Miller's ominous comment flagged by ex-Trump aide: 'Read into that what you want'
Stephen Miller's ominous comment flagged by ex-Trump aide: 'Read into that what you want'
Former Department of Homeland Security chief of staff Miles Taylor has become one of President Donald Trump's targets after spending the better part of five years speaking out. Now he's taking another big step.Sarah K. Burris (Raw Story)
adhocfungus likes this.
Mushrooms May Replace Metal in Future Computers — And You Could Build One At Home
Mushrooms May Replace Metal in Future Computers — And You Could Build One At Home
Learn how scientists found a way to turn mushrooms into computer chips and how these living computers can make the future of technology more sustainable.Stephanie Edwards (Discover Magazine)
adhocfungus likes this.
'Keep Android Open' movement fights back against Google sideloading restrictions
'Keep Android Open' movement fights back against Google sideloading restrictions
: Petition seeks to rally community opposition and alert regulatorsThomas Claburn (The Register)
like this
adhocfungus, copymyjalopy e essell like this.
Europe’s rights court clears Norway of climate misconduct over Arctic oil licences
Europe’s top human rights court ruled Tuesday that Norway did not violate its climate obligations by granting Arctic oil and gas exploration licenses in 2016. It was a setback for climate activists after a landmark ruling last year against Switzerland for failing to take sufficient action on climate change.
UK far right movement is still led by anonymous accounts
But an analysis of social media suggests something else. Many people and groups on the radical and far-right are harnessing a process known as audience capture in order to influence political policy.A group of anonymous X accounts is said to follow a “posting-to-policy” strategy. These accounts – some of which are run by disaffected Westminster professionals – post to inject their grievances into online discourse.
To explore this dynamic, and how Reform’s recent u-turn has been shaped by it, we analysed the online networks that drove conversation about “mass deportations” on X over the past year. Using computational methods, we identified four distinct sub-communities defined by their retweet relationships. These sub-communities were formed around far-right influencers, radical right influencers, Advance UK/free-marketeer influencers – and around the Reform party.
Discussion of mass deportations in 2024 was almost exclusively dominated by the far-right and the anonymous accounts of the radical right. Fast forward to April 2025 and we find Lowe, Habib and a wider range of rightwing influencers have entered the conversation in support of the policy.
Finally, in September, following Reform’s August announcement, you can see Farage and key Reform personnel supplant the influencers as players in a movement they had little role in creating. In doing so, the party has aligned itself with a policy that less than a year ago it vehemently rejected.
Inside the far-right social media ecosystem normalising extremist ideas in UK politics
A process of normalisation has led Reform to propose mass deportations where once it believed such a policy would never be politically viable.The Conversation
Amanhã a babá não trabalha
Crédito: Tomaz Silva/Agência Brasil
A FAVELA SANGRA, E O PAÍS FINGE NORMALIDADENas coberturas à beira–mar, a vida segue, porque o sangue derramado não mancha o piso de mármore
AOL sold to Bending Spoons for $1.5 Billion
Bending Spoons, the company behind Evernote, Meetup, StreamYard and WeTransfer is acquiring AOL from Yahoo!'s new owner, Apollo Global Management, for $1.5 Billion.
the-independent.com/news/world… (Archive]
You’ve got a new owner: AOL is sold in reputed $1.5B deal to tech conglomerate
AOL, which became symbolic of the early internet age, is being bought by Italian tech company Bending SpoonErin Keller (The Independent)
What happens if Antifa is labeled a foreign terrorist organization
The administration wants to label antifata as a foreign terrorist organization. This would add legal teeth to stop anti fascists. Any organization that this administration thinks is anti nazi or anti fascist could legally be shutdown if this goes through.
like this
adhocfungus e copymyjalopy like this.
like this
adhocfungus likes this.
[Video] "I Tried the First Humanoid Home Robot. It Got Weird."
reshared this
Technology reshared this.
Israeli soldier denied entry into Prague — airport detains murderer
Authorities at Prague airport detained an Israeli soldier who actively helped the terrorist state commit genocide. Israeli soldier denied entry into the Czech Republic after a 15-hour detention.
Great Job PragueAn Israeli Soldier who fought in Gaza and Lebanon was detained for hours at Prague’s international airport and denied entry to the Czech Republic. pic.twitter.com/yRI6E6vHwG
— Ryan Rozbiani (@RyanRozbiani) October 29, 2025
According to ynet Global, the man travelled to Prague with his wife “after months of reserve duty.” Or, in other words, after months of murdering innocent Palestinians. Hilariously, he also claimed they ‘treated him like a criminal.’ Funny that.
In the end, the couple had to pay for their own return tickets and lost their holiday.
Reports suggest he was denied entry to the Schengen Area after French authorities issued a well-earned criminal alert against him in the Schengen Information System. Authorities put them through 15 hours of questioning before deportation to Israel.
Israeli soldier denied entry: No safe haven for war criminals.
He claimed:
I don’t understand why we’re being deported or what I supposedly did that led to this kind of ‘warning’ against me.
These people are so far removed from reality that they don’t realise that committing genocide might raise the occasional red flag when travelling internationally.
this is how we must respond to all them genocidal cunts who get off on killing innocent people. exclude them, make it known that they are not welcome anywhere. t.co/h8x38et9ve— ف (@jiminilvrs) October 29, 2025
The Schengen zone prevents travel for individuals with certain criminal records, including drug trafficking and murder. While the Schengen criteria don’t explicitly include ‘war crimes’ or ‘genocide’ in black and white. But anyone with a brain can put two and two together.
He also tried to suggest that someone had stolen his identity and used it to commit serious crimes. Again, did murdering babies slip his mind?
Good, I hope other countries follow suit t.co/ElpospykYf— MillieMN001 (@MillieMN001) October 29, 2025
May this be the beginning of a worldwide response. No safe haven for war criminals. t.co/GL3AiMVeMJ
— Jugni (@kikigee24) October 29, 2025
Earlier this month, British lawyers said that British courts can now try and jail Brits who served in Israel’s genocide, under the Foreign Enlistment Act. This is thanks to the UK finally recognising the Palestinian state. However, the law cannot be applied retrospectively. This means that those who have served in the IDF over the two years prior to September will effectively get away scot-free.
Declassified UK have previously reported that 80 Brits were serving in the Israeli military on 7 October 2023. This raises questions about a future in which the UK will have baby-murdering ex-IDF soldiers roaming the streets.
One of these days we will be arresting them for their warcrimes and putting them on trial. t.co/ulAQ14kONt— Dianne Woodward (@WoodwarddianneJ) October 29, 2025
It’s sickening that IDF soldiers feel such entitlement for their little autumn holiday, after trapping 2 million Palestinians in Gaza while they’ve besieged it into oblivion
Now, if a certain British prime minister could stop inviting violent Israeli thugs and war criminals into the UK, that would be great.
Isaac Herzog should be investigated for war crimes during UK visit
Diplomatic immunity should not stop police from interviewing Isaac Herzog and otherwise investigating him over war crimesThe Canary
How comprimised is Ed Zitron (Where's Your Ed At)?
You need to use the tools of the job you've chosen to do
Web dev at the end of the world, from Hveragerði, Icelandwww.baldurbjarnason.com
The Insanity of the Facebook Puzzle Scam Code: “BE CV BK 2025 -R-D” and the Unbelievable Spread of an Obvious Scam
It’s hard to overstate just how bizarre it is that something as nonsensical as “BE CV BK 2025 -R-D” has taken over Facebook and even started creeping into Google search results. This strange code — which looks like some mix of a fake model number, a coded message, and a bot gibberish tag — has appeared in thousands of posts across Facebook. And what’s wild is that, despite being so obviously a scam, so clearly fraudulent, so transparently fake, it’s everywhere. The fact that it’s not being widely discussed, not being reported on by major outlets, not being taken down effectively by Facebook, makes the whole thing even more insane.
You can go on Facebook right now, type that code into the search bar — “BE CV BK 2025 -R-D” or “BE CV BK.2025 -R-D” — and what you’ll find is a flood of the same kinds of posts. Some are in different languages. Some use emojis. Some pretend to be part of “puzzle groups” or “mystery challenges.” Others are just random accounts spamming the same text over and over again, often accompanied by weird links, grainy photos, or random “game” announcements. But the one thing they all share is the same exact scam code.
The strangest part is that this isn’t just some obscure niche spam chain buried deep in Facebook’s murky corners. It’s out in the open. Public groups. Public pages. Public posts. You can find it by simply searching. It’s like the digital equivalent of walking through a city and seeing “SCAM” graffiti plastered across every wall — and somehow, no one’s talking about it.
That’s what makes this whole “puzzle scam” phenomenon feel so surreal. It’s not hidden. It’s not subtle. It’s right there in plain sight. And yet, despite being so blatant, it’s spreading like wildfire.
It’s easy to see why the “puzzle” angle works. These kinds of scams often rely on curiosity — on the human desire to “figure out” something mysterious. The code looks cryptic enough to seem like there’s a deeper meaning behind it. “BE CV BK 2025 -R-D.” It almost feels like it could be a secret message, or a part of a viral challenge, or some kind of ARG (alternate reality game). And that’s what hooks people in. Someone sees a friend post it. They think, “What is this? Is this some new Facebook game? Is this part of something?” And before long, they’re clicking links, joining groups, following instructions, or even sharing the post themselves — unknowingly helping to spread the scam further.
The entire design of this “puzzle” is meant to exploit one of the simplest psychological triggers: curiosity. Humans are hardwired to seek answers, especially when something looks like a code or a mystery. Scammers have known this for years — that’s why “riddles,” “tests,” “IQ puzzles,” and “hidden messages” have long been a popular front for phishing scams, malware links, and data-harvesting schemes. This particular Facebook scam just takes that formula and dresses it up with a meaningless code that looks intriguing to the untrained eye.
But what’s really unsettling about this whole thing is just how many posts there are. It’s not just a handful of scammers copying and pasting the same message. There are thousands. Some of them are weeks or months old. Others are being posted in real time. The scam has evolved into a kind of bot swarm, almost like a virus that keeps replicating itself across the platform. And the lack of any large-scale intervention from Facebook makes it even worse.
You’d think a platform with as much power, as much data control, and as much AI filtering as Facebook would be able to catch something as blatantly repetitive and nonsensical as this. But nope. The scam lives on, thriving. And that’s what’s disturbing. The scammers have found a way to stay one step ahead — maybe by slightly changing punctuation, or spacing, or formatting, to keep slipping past Facebook’s algorithmic filters. The difference between “BE CV BK 2025 -R-D” and “BE CV BK.2025 -R-D” might be enough to fool automated moderation systems.
And meanwhile, the rest of us are just sitting here, watching this nonsense flood our feeds, while hardly anyone seems to be calling it out.
It’s a sign of how desensitized we’ve all become to online spam. There’s so much garbage on the internet — from fake giveaways to impersonation accounts to AI-generated comment bots — that something like this barely registers anymore. The absurdity of a code like “BE CV BK 2025 -R-D” showing up everywhere doesn’t even faze people anymore. We’ve reached a point where mass spam has become so normalized that people just scroll past it without question.
But the danger here isn’t just about annoyance. It’s about what’s behind these scams. Many of these “puzzle” posts are actually phishing attempts or clickbait traps that redirect users to shady sites. Others use the puzzle format to get users to comment, share, or click a “Continue” button — all tactics designed to collect engagement data or personal information. And then there’s the possibility that some of these are part of larger coordinated bot networks — networks designed not just to scam individuals, but to manipulate engagement metrics, artificially inflate content visibility, or even test out new spam strategies that can later be used in political or commercial manipulation.
That may sound far-fetched, but it’s not. Facebook has long been a testing ground for disinformation and bot campaigns. If scammers can flood the platform with something so meaningless yet widespread, imagine what they can do when they actually put some effort into it.
What’s also strange is how the scam has spread to Google. Search “BE CV BK 2025 -R-D” and you’ll see that it’s indexed in all kinds of pages — cached Facebook links, random blog comment sections, obscure reposting sites. The digital footprint of this nonsense code is massive. And that means it’s not just a Facebook issue anymore. It’s become part of the broader web ecosystem, another layer in the weird, polluted strata of modern internet junk data.
It’s almost poetic, in a depressing way. The internet used to be about connection, creativity, and genuine curiosity. Now that same curiosity — the thing that once drove people to explore and learn — is being weaponized against them. Instead of solving puzzles for fun, people are being tricked into interacting with spam. Instead of decoding art or mystery, they’re decoding scams. And it’s not even subtle anymore.
What’s wild, too, is that Facebook users themselves are often the ones unknowingly keeping it alive. The bots can only do so much — but when real people start engaging, commenting, sharing, or trying to “warn” others by reposting the code, that activity actually boosts the visibility of the scam. Facebook’s algorithm doesn’t care why something is getting engagement — it just sees numbers. So every time someone posts, “Don’t fall for BE CV BK 2025 -R-D, it’s a scam!”, that post can ironically push the code further up the visibility ladder, leading even more people to see it.
The whole thing feels like an ouroboros of internet stupidity — a self-feeding loop where spam generates attention, attention generates engagement, and engagement keeps the spam alive.
And maybe that’s the most disturbing part of all: how effortless it’s become for something like this to go viral without any real content behind it. It doesn’t even have to make sense. It doesn’t have to be convincing. It doesn’t have to look real. It just has to exist in large enough quantity to trick the algorithm.
It’s a perfect reflection of how broken online ecosystems have become. In the old internet, scams had to at least try to look legitimate — a fake website pretending to be your bank, or a phony giveaway with a convincing logo. Now? All it takes is a random string of letters and numbers, a few thousand bot accounts, and a platform too busy or too lazy to do anything about it.
Facebook’s failure to stop something this blatant speaks volumes. It’s not just an oversight — it’s a sign that their moderation systems are reactive, not proactive. They’re so focused on surface-level metrics that something like this can thrive indefinitely. And in that sense, the “BE CV BK 2025 -R-D” code becomes more than just a scam. It becomes a symptom. A sign of decay. Proof that the systems that were supposed to protect users from obvious manipulation are no longer functioning as intended.
It’s worth asking: what’s the endgame here? What’s the point of this code? Is it just engagement farming? A front for phishing? A bot experiment? Or is it something even weirder — an automated system left to run amok, spamming for the sake of spamming?
At this point, no one really knows. But that’s the scary part — no one’s really trying to find out, either. The internet is so overloaded with noise that even something this widespread can go largely unnoticed by the mainstream. People see it, shrug, and move on.
That’s how scams survive. Not because they’re convincing, but because people have stopped caring enough to investigate.
Maybe that’s the biggest takeaway from the “BE CV BK 2025 -R-D” puzzle scam — not just how it spreads, but what it reveals about us. We’re living in a time where nonsense thrives because attention is cheap. Where scams succeed not through sophistication, but through sheer saturation. Where even the most absurd, poorly disguised fraud can blanket an entire social network and nobody blinks.
The “BE CV BK 2025 -R-D” code isn’t just a scam — it’s a mirror. A reflection of an online culture that’s too burned out, too overwhelmed, and too desensitized to call out the obvious anymore.
And maybe, until more people start noticing the sheer absurdity of things like this, we’re going to keep seeing the same pattern play out — again and again — until our feeds are nothing but codes, spam, and empty noise pretending to be meaning.
Also on:
Bridgy Fed
Bridgy Fed is a bridge between decentralized social networks like the fediverse, Bluesky, and web sites and blogs.fed.brid.gy
Emergent introspective awareness in large language models
Emergent introspective awareness in large language models
Research from Anthropic on the ability of large language models to introspectwww.anthropic.com
reshared this
Technology reshared this.
I tend to see a lot of discussion taking place on here that's pretty out of touch with the present state of things, echoing earlier beliefs about LLM limitations like "they only predict the next token" and other things that have already been falsified.
This most recent research from Anthropic confirms a lot of things that have been shifting in the most recent generation of models in ways that many here might find unexpected, especially given the popular assumptions.
Specifically interesting are the emergent capabilities of being self-aware of injected control vectors or being able to silently think of a concept so it triggers the appropriate feature vectors even though it isn't actually ending up in the tokens.
They aren't "self-aware" at all. These thinking models spend a lot of tokens coming up with chains of reasoning. They focus on the reasoning first, and their reasoning primes the context.
Like if I asked you to compute the area of a rectangle you might first say to yourself: "okay. There's a formula for that. LxW. This rectangle is 4 by 5, so the calculation is 4x5, which is 20." They use tokens to delineate the "thinking" from their response and only give you the response, but most will also show the thinking if you want.
In contrast, if you ask an AI how it arrived at an answer after it gives it, it needs to either have the thinking in context or it is 100% bullshitting you. The reason injecting a thought affects the output is because that injected thought goes into the context. It's like if you're trying to count cash and I shout numbers at you, you might keep your focus on the task or the numbers might throw off your response.
Literally all LLMs do is predict tokens, but we've gotten pretty good at finding more clever ways to do it. Most of the advancements in capabilities have been very predictable. I had a crude google augmented context before ChatGPT released browsing capabilities, for instance. Tool use is just low randomness, high confidence, model that the wrapper uses to generate shell commands that it then runs. That's why you can ask it to do a task 100 times and it'll execute 99 times correctly and then fail—got a bad generation.
My point is we are finding very smart ways of using this token prediction, but in the end that's all it is. And something many researchers shockingly fail to grasp is that by putting anything into context—even a question—you are biasing the output. It simply predicts how it should respond to the question based on what is in its context. That is not at all the same thing as answering a question based on introspection or self-awareness. And that's obviously the case because their technique only "succeeds" 20% of the time.
I'm not a researcher. But I keep coming across research like this and it's a little disconcerting that the people inventing this shit sometimes understand less about it than I do. Don't get me wrong, I know they have way smarter people than me, but anyone who just asks LLMs questions and calls themselves a researcher is fucking kidding.
I use AI all the time. I think it's a great tool and I'm investing a lot of my own time into developing tools for my own use. But it's a bullshit machine that just happens to spit out useful bullshit, and people are desperate for it to have a deeper meaning. It... doesn't.
So while your understanding is better than a lot of people on here, a few things to correct.
First off, this research isn't being done on the models in reasoning mode, but in direct inference. So there's no CoT tokens at all.
The injection is not of any tokens, but of control vectors. Basically it's a vector which being added to the activations makes the model more likely to think of that concept. The most famous was "Golden Gate Claude" that had the activation for the Golden Gate Bridge increased so it was the only thing the model would talk about.
So, if we dive into the details a bit more…
If your theory was correct, then the way the research asks the question saying that there's control vectors and they are testing if they are activated, then the model should be biased to sometimes say "yes, I can feel the control vector." And yes, in older or base models that's what we might expect to see.
But, in Opus 4/4.1, when the vector was not added, they said they could detect a vector… 0% of the time! So the control group had enough introspection capability as to not stochastically answer that there was a vector present when there wasn't.
But then, when they added the vector at certain layer depths, the model was often able to detect that there was a vector activated, and further to guess what the vector was adding.
So again — no reasoning tokens present, and the experiment had control and experimental groups where the results negates your theory as to the premise of the question causing affirmative bias.
Again, the actual research is right there a click away, and given your baseline understanding at present, you might benefit and learn a lot from actually reading it.
I think we could have a fascinating discussion about this offline. But in short here's my understanding: they look at a bunch of queries and try to deduce the vector that represents a particular idea—like let's say "sphere". So then without changing the prompt, they inject that concept.
How does this injection take place?
I played with a service a few years ago where we could upload a corpus of text and from it train a "prefix" that would be sent along with every prompt, "steering" the output ostensibly to be more like the corpus. I found the influence to be undetectably subtle on that model, but that sounds a lot like what is going on here. And if that's not it then I don't really follow exactly what they are doing.
Anyway my point is, that concept of a sphere is still going into the context mathematically even if it isn't in the prompt text. And that concept influences the output—which is entirely the point, of course.
None of that part is introspective at all. The introspection claim seems to come from unprompted output such as "round things are really on my mind." To my way of thinking, that sounds like a model trying to bridge the gap between its answer and the influence. Like showing me a Rorschach blot and asking me about work and suddenly I'm describing things using words like fluttering and petals and honey and I'm like "weird that I'm making work sound like a flower garden."
And then they do the classic "why did you give that answer" which naturally produces bullshit—which they at least acknowledge awareness of—and I'm just not sure the output of that is ever useful.
Anyway, I could go on at length, but this is more speculation than fact and a dialog would be a better format. This sounds a lot like researchers anthropomorphizing math by conflating it with thinking, and I don't find it all that compelling.
That said, I see analogs in human thought and I expect some of our own mechanisms may be reflected in LLM models more than we'd like to think. We also make decisions on words and actions based on instinct (a sort of concept injection) and we can also be "prefixed" for example by showing a phrase over top of an image to prime how we think about those words. I think there are fascinating things that can be learned about our own thought processes here, but ultimately I don't see any signs of introspection—at least not in the way I think the word is commonly understood. You can't really have meta-thoughts when you can't actually think.
Shit, this still turned out to be about 5x as long as I intended. This wasn't "in short" at all. Is that inspection or just explaining the discrepancy between my initial words and where I've arrived?
The injection is the activation of a steering vector (extracted as discussed in the methodology section) and not a token prefix, but yes, it's a mathematical representation of the concept, so let's build from there.
Control group:
Told that they are testing if injected vectors present and to self-report. No vectors activated. Zero self reports of vectors activated.
Experimental group:
Same setup, but now vectors activated. A significant number of times, the model explicitly says they can tell a vector is activated (which it never did when the vector was not activated). Crucially, this is only graded as introspection if the model mentions they can tell the vector is activated before mentioning the concept, so it can't just be a context-aware rationalization of why they said a random concept.
More clear? Again, the paper gives examples of the responses if you want to take a look at how they are structured, and to see that the model is self-reporting the vector activation before mentioning what it's about.
I've read it all twice. Once a deep skim and a second more thorough read before my last post.
I just don't agree that this shows what they think it does. Now I'm not dumb, but maybe it's a me issue. I'll check with some folks who know more than me and see if something stands out to them.
None of this obfuscation and word salad demonstrates that a machine is self-aware or introspective.
It's the same old bullshit that these grifters have been pumping out for years now.
morgunkorn
in reply to armony • • •You could pay me and i still wouldn't want to use any of this.
I can write my own texts, I can read long ones without having to get a summary. I can draw, I can take pictures, I can do online research. All by myself without a spicy autocomplete to prechew it for me.