Kremlin expects no ‘magical breakthroughs’ at talks
Kremlin expects no ‘magical breakthroughs’ at talks
Dmitry Peskov emphasized that Russia intended to continue pursuing its interests and to fulfill the tasks that had initially been setTASS
like this
originalucifer likes this.
Dell confirms breach of test lab platform by World Leaks extortion group
A newly rebranded extortion gang known as "World Leaks" breached one of Dell's product demonstration platforms earlier this month and is now trying to extort the company into paying a ransom.
Ex-officer sentenced to 33 months in prison in Breonna Taylor case
Brett Hankison, a former Kentucky police officer who was convicted in the death of Breonna Taylor, a 26-year-old emergency medical technician, was sentenced on Monday to 33 months in prison.
Taylor was shot and killed on March 13, 2020, during a botched drug raid authorized by the Louisville Metro Police Department. A Louisville detective at the time, Hankison, 46, was found guilty last November of violating Taylor's civil rights while executing a search warrant on her home, which resulted in the tragedy.
Hankison will not report directly to prison, with U.S. District Judge Rebecca Grady Jennings saying during Monday's sentencing hearing that the Bureau of Prisons will decide when his sentence begins, according to The Associated Press. His prison sentence will be followed by three years of supervised probation.
☆ Yσɠƚԋσʂ ☆ likes this.
Tesla is the least-trusted car brand in America
Self-Driving Cars and Electric Vehicles: U.S. Market Insights & Analysis, July - Electric Vehicle
EVIR surveyed more than 8,000 U.S. consumers, weighted by education, race, gender, age, income, and geography, to uncover the truths behind what’s powering or stalling the EV and self-driving car transition.sedrick@kineticstrategies.us (ev-intelligence.com)
copymyjalopy likes this.
I cant be the only one that sucks at playing against humans but am OK against bots???
essell likes this.
Sweden, Sex Work, Screens: The Criminalisation of Online Sex Work and Article 8 of the ECHR
Sweden has quietly taken a radical step: it is now illegal to purchase online sexual acts. This move advances Sweden’s long-standing “end demand” policy model for tackling sexual services from the physical realm, into the digital. Yet it seems to overlook the significant differences between the two spheres – in terms of behaviour models, profiles, and market dynamics – and how such differences may be taken into account when determining the persuasiveness of the law’s rationale. This becomes especially clear when measured against the protections enshrined under Article 8 of the European Convention on Human Rights (ECHR) and recent Strasbourg case law.While the criminalisation of the purchase of in-person sexual services has been judged to be compatible with Article 8, the underlying reasoning rests on factors that do not translate to the online sphere: combatting prostitution and human trafficking, a lack of consensus on sex work policy across Europe, and an inability to parse the harms caused by the law from the harms caused by sex work itself. Sweden’s extension of its “end demand” policy into digital sex work thus risks overstepping the boundaries of Article 8 of the ECHR and reveals how laws that are directly transplanted from the offline to the online sphere without due thought may lead to the erosion of private digital rights.
Lavrov Issues Strong Warning Against Goading Ukraine Into Acts of Terrorism
Lavrov Issues Strong Warning Against Goading Ukraine Into Acts of Terrorism
All attempts to incite Kiev to commit terrorist acts will receive an adequate response, Moscow will achieve all its goals, Russian Foreign Minister Sergey Lavrov said on Tuesday.Sputnik International
Meta Censored Critical Updates on Syria’s Regime Change, Prompting Oversight Board Review
Meta Censored Critical Updates on Syria’s Regime Change, Prompting Oversight Board Review - SMEX
In December 2024, Syrians witnessed the unthinkable: the collapse of Assad’s regime by a coalition of armed groups, led by...Yasmina Zein (SMEX)
like this
thisisbutaname, Maeve e EpicFailGuy like this.
Technology reshared this.
It would be nice if you could post something where we can examine the source. (EDIT: the link has been changed since I wrote this)
I found this article: techspot.com/news/108720-hidde…
There they say that it's not yet ready to be used in evidence, but the problem with that is that most forensic "science" is generally misapplied and nowhere near as conclusive as the police want us to think. They can usually massage the results to tell a jury what they want to be true. That would be my concern with this kind of technique.
Also, if you're going to the trouble of making a 3d printed ghost gun that will be used in a crime, you could always hide the toolmarks with a sander. You could also treat the surface with resin which would make the markings practically unrecoverable. I've started doing both of these for my prints and I love the results just for the aesthetics, so it's not such a stretch to imagine a gunsmith doing the same.
The hidden fingerprints inside 3D-printed ghost guns
Kirk Garrison, a forensics expert with the San Bernardino Sheriff's Department, has spent years working at the intersection of digital blueprints, heated plastic filaments, and real-life criminal...Skye Jacobs (TechSpot)
like this
riot likes this.
There they say that it’s not yet ready to be used in evidence, but the problem with that is that most forensic “science” is generally misapplied and nowhere near as conclusive as the police want us to think.
This is such an important thing to remember. It's just like how a lot of hand-held breathalyzers are closed source, and when their source code is finally subjected to scrutiny (because you're supposed to be able to face your accuser, and the device is your accuser), it often doesn't meet basic required standards for things like error reporting or failsafes to prevent false positives.
Much of forensic "science" isn't exactly science as we understand it.
Yup, Behind the Bastards did an excellent two parter on forensic science in general:
iheart.com/podcast/105-behind-…
iheart.com/podcast/105-behind-…
They make a good point that real science is involved, but by the time it makes it into the police's hands it's mutated into essentially a mechanism to manufacture convictions. Grifters get hold of the science, and cops are like the perfect marks, because they're just primed for anything that will confirm their existing biases, plus they've got massive state budgets to play with, and they'll happily give the grifters legitimacy.
Part One: The Bastards of Forensic Science - Behind the Bastards | iHeart
Forensic Science is supposed to provide perfect certainty in the most serious criminal cases. What if it's all a bunch of bullshit? Robert sits down with Dr. Kaveh Hoda to talk about all the myriad cons in forensic 'science.”See omnystudio.iHeart
Kirk Garrison, a forensics expert who works for the San Bernardino Sheriff’s department, told 404 Media he’s had early success matching 3D printed objects to the machines that made them.
This is "bite evidence" all over again, isn't it? For those not familiar, cops swore in court they could match a perp's teeth to bite marks on victim's bodies.
They couldn't.
There were a lot of tainted court cases because of their junk science. I'm all for murderers going to prison but lets not use bullshit to lock up perhaps the wrong people.
like this
Maeve e giantpaper like this.
Just based on the title, since I haven't read the article yet, that's quite unfortunate. I would want all of them to be practically identical, so there's no way to trace them at all.
Edit: After reading the article, I stand by my above statement. I still want them to be completely identical if at all possible.
Some good discussion from /c/3dprinting@lemmy.world about how it really isn’t as dependable/tracable as the article says:
TLDR is 3D printing typically uses brass nozzles which wear down over time which will change marks left over time, your bed leveling can change over time, and the “fingerprints” the article talks about can be avoided by just printing in different positions/rotations on the bed.
like this
Maeve likes this.
like this
Maeve likes this.
Show me 20 people with 3d printers and I’ll show you 20 other people with 3d printers that match the fingerprints of the first 20.
This isn’t like paper printers where companies were forced by the government to encode the serial numbers of the printer into every piece of paper that comes out. There’s no way you could hide identifying information in molten plastic like that.
US withdraws from UNESCO over ‘anti-Israel bias’
US withdraws from UNESCO over ‘anti-Israel bias’
This is the second time Washington has withdrawn from UNESCO since 2011, following the agency’s decision to admit Palestine as a member statethecradle.co
like this
geneva_convenience likes this.
like this
geneva_convenience likes this.
How can I share/store sensitive data for family
I need to start making plans for when I am gone, much sooner than I thought, and I realized our finances are pretty opaque to my spouse. Our bank account is shared, but there are other sites that only I have access to.
The easiest solution would be to physically write down logins and what needs done, put it in an envelope, and tell my family where that envelope is. I'm not thrilled about that, because I would have to shred and rewrite it every time I update a password or a URL changes, and it'd be vulnerable to nosy guests.
Putting it in a shared Google Doc would be easiest for everyone. But then Google has that data. Even supposing I trust a cloud SaaS provider not to misuse the data (which is a big 'if') I do not trust them to never have a data breach.
Self-hosting seems like the next step, except I expect my home server to be the first thing to collapse once I'm gone. Filing login info with an estate attorney would still require frequent updates. Putting a document on a flash drive risks data loss, but is what I'm leaning towards.
Is there a solution I'm missing?
like this
giantpaper likes this.
like this
adhocfungus likes this.
like this
adhocfungus likes this.
like this
adhocfungus likes this.
Password manager such as Bitwarden, you can store your passwords, and sensitive info as notes or attachments. It's all encrypted client side.
Then you just need to have a note with the master password and instructions on how to access it.
like this
adhocfungus likes this.
like this
giantpaper likes this.
every time I update a password
Use a password manager. KeePassXC stores stuff in a file, so it's easier to synchronize. You can selfhost BitWarden too.
Syncthing is great to synchronize stuff across devices.
Cryptomator creates encrypted volumes (looks like a folder with gibberish inside) for you, which you can sync with whatever commercial cloud.
Putting a document on a flash drive
Data loss might come from bitrot, yes. Regardless, you should always have multiple backups.
Self-host Bitwarden | Bitwarden
Customers wishing to self-host a Bitwarden server for their organization or personal use have a variety of deployment options, including options for the server and infrastructure Bitwarden is deployed on, the database used by the server, and the cert…Bitwarden
like this
adhocfungus likes this.
This is the one reason I’ve paid for 1Password. My wife has access and can get what is needed without figuring out how to revive a self hosted password solution. I realize this isn’t about self hosting, and that you can pay for Bitwarden too. It just struck a chord.
OP wishing you all the best.
like this
adhocfungus e giantpaper like this.
Why not just put logins in a database such as keepass and then have the password for that written down in like a lock box or something?
You could also store a flash drive with the password in the lock box and update it, say, every six months with the most current database version.
like this
adhocfungus e giantpaper like this.
No solution I've found, but I've been working on this myself. As I see it, there are two situations, and four categories of data:
I. My wife survives me
II. We both die, e.g. in a car
- Digital media
- Financial accounts
- Subscriptions
- Physical possessions
I've been thinking about getting an M-Disk writer for media, because ultimately, backing up to B2 is fine until I'm gone. Family members will need physical media for the photos and stuff.
For secrets, I'm planning using SSSS. Keys will be given to members on each side of my wife's and my families. If we both die, they'll have to get together, put their keys together, and decrypt the KeePass DB.
The online accounts are almost all financial; those are in a KeePass DB. My wife already has access to all of that through power is attorney, and if we both go, it's SSSS for the family.
The third data category are accounts and services that will be to be stopped. I don't subscribe to much, but the VPS provider and B2 will have to be terminated, and a document with instructions and with the credentials is in the SSSS archive.
The final category are assets: home, mortgage info, where and what the M-Disks are, a copy of the will that deals with all of the valuables, and any notes about anything not covered in the will. That's in documents in the SSSS archive.
I still have to put the archive together. I've been working toward a state where all of the secrets are in a cryptfs that's shared on the LAN and automatically encrypted with SSSS and synced to a share. Once I have that automated, I'll communicate out the SSSS keys and a how-to document.
In some ways, it was easier when you just died and your kids fought over the china. But I have a plan.
My mother died recently, and she was the breadwinner and was in charge of everything financial, because my surviving father is a toxic narcissist with zero financial literacy who refuses help from anyone. So I just have to say kudos to you for thinking about this difficult stuff. Your family will appreciate it more than you can imagine.
Other commenters have already given you solid advice, and I don’t have anything to add there, but more people need to have these difficult conversations and make real practical preparations, as soon as possible. Speaking from experience, not having clear guidance about where things are and what should be done with them, makes an already emotional situation even harder to deal with. Everybody dies, but in death you can make your family’s grieving process slightly easier by thinking ahead like this.
I’m sorry for whatever you’re going through, but props for thinking about other people while you go through it.
like this
giantpaper e adhocfungus like this.
adhocfungus likes this.
"In this envelope is the password for my keepass password vault. The entry for "In case of emergency" contains everything you should need to know in the event of my incapacitation or worse.
There are two USB keys with this vault on them, they are synchronized for redundancy. When I pass, get the password out of this envelope, plug in a USB key, open keypass and enter the password. "
You: Use the primary key as your password storage, keep the backup key plugged into a raspberry pi, run syncthing on both devices.
Have a spare test key set up, do a dry run with the family members you entrust to have this data.
like this
adhocfungus likes this.
like this
adhocfungus likes this.
adhocfungus likes this.
Filen – Next Generation End-To-End Encrypted Cloud Storage
Filen – Next Generation End-To-End Encrypted Cloud Storage. Get started with 10 GB of free space.filen.io
adhocfungus likes this.
You kinda only need the email credentials. Shouldn't the rest be resettable from that point?
Is there anything that needs MFA that they won't have?
like this
adhocfungus likes this.
adhocfungus doesn't like this.
I would use Keepass. You would have a single file, opened with a single password, that you could share with them however you want.
Wishing you the best
adhocfungus likes this.
adhocfungus doesn't like this.
Can LLMs Do Accounting? Evaluating LLMs on Real Long-Horizon Business Tasks
Can LLMs Do Accounting? | Penrose
An experiment exploring whether frontier models can close the books for a real SaaS company.accounting.penrose.com
I do not believe that LLMs will ever be able to replace humans in tasks designed for humans. The reason is that human tasks require tacit knowledge (=job experience) and that stuff is not written down in training material.
However, we will start to have tasks for LLMs pretty soon. It was already observed that LLMs work better on stuff produced by other LLMs.
To be fair, not all knowledge of LLM comes from training material. The other way is to provide context to instructions.
I can imagine someone someday develops a decent way for LLMs to write down their mistakes in database and some clever way to recall most relevant memories when needed.
GitHub - MemTensor/MemOS: MemOS (Preview) | Intelligence Begins with Memory
MemOS (Preview) | Intelligence Begins with Memory. Contribute to MemTensor/MemOS development by creating an account on GitHub.GitHub
You sort of described RAG. It can improve alignment, but the training is hard to overcome.
See Grok that bounces from “woke” results to “full nazi” without hitting the mid point desired by Musk.
You're describing neurosymbolic AI, a combination of machine learning and neural network (LLM) models. Gary Marcus wrote an excellent article on it recently that I recommend giving a read, How o3 and Grok 4 Accidentally Vindicated Neurosymbolic AI.
The primary issue I see here is that you're still relying on the LLM to reasonably understand and invoke the ML models. It needs to parse the data and understand what's important in order to feed it into the ML models and as has been stated many times, LLMs do not truly "understand" anything, they are inferring things statistically. I still do not trust them to be statistically accurate and perform without error.
How o3 and Grok 4 Accidentally Vindicated Neurosymbolic AI
Neurosymbolic AI is quietly winning. Here’s what that means – and why it took so longGary Marcus (Marcus on AI)
Compounding tasks like accounting where operations sound easy but create a chain of counter entries and balances need to be organized by account is none of AIs business until they can prove that multiple sequential steps can have over 99% accuracy and the checksum of the accounts is balanced.
Multiple sequential steps with six operations where we assume 99 PCT of each, is right can equal 90% accuracy.
This also reads at least 10% errors if they all six go wrong.
How many different entries are in a company over a month's closure?
Now this wouldn't be and issue if it can balance the statements of consolidated accounts and find where are we missing entries or misallocations. That sir is why we pay someone with experience.
I quit my job in public accounting for many reasons, but the primary one was the forceful adoption of LLMs to replace associates.
I told the dimwits at the top that it was a mistake, because LLMs are incompetent even when the information fed to it was perfect, and that was rarely the case in practice.
Our ultra wealthy clients were notorious for giving us the most incomplete and asinine information, and it often took someone with decades of experience to decipher what the fuck their personal assistants are even talking about.
They went ahead anyway because of the high cost of wages, of course, and I made my exit because I did not wish to be complicit in such a monumental mistake.
Lmfao the LLM they laid associates off and paid half a million dollars for made up fake ledger accounts when accounts didn't reconcile, and none of the dumbasses left noticed in time because they hadn't done associate-level work in decades.
It also lied all the time, even when you asked it not to.
The damage was done and the biggest clients started leaving, so they begged us all to come back but I got obsessed with baking bread and I ain't about to neglect my sourdough starters to help a group of people who would lose a battle of wits against yeast.
reshared this
Kent Navalesi ☕️ reshared this.
like this
EpicFailGuy e Endymion_Mallorn like this.
Technology reshared this.
But then he would have been a good human, and one who worried about the right things.
You can’t just swap suicide like that.
UN Statements Undercut New Israeli Report on 10/7 Sexual Violence
Major news organizations, most prominently the New York Times, have promoted the idea of systematic sexual violence at opportune moments to justify Israel’s ongoing genocide in Gaza. The first major salacious headlines and assertions emerged in late 2023, when Israel was campaigning to restart its killing during a brief ceasefire. The latest effort to revive this narrative follows the same pattern as its predecessors—and, indeed, is more overtly political, with the report spending less airtime on the well-being of women than on reasons we should roll back what is left of international law.
The UN, however, has stated multiple times that it does not have evidence of systematic sexual abuse by Hamas or any other militant group on October 7, 2023. A top United Nations official issued a statement last week that stands in direct contradiction to the new Israeli report.
Reem Alsalem, the UN Special Rapporteur on violence against women and girls, affirmed in her statement this week that though the UN had not found “systematic” sexual violence: "It is my understanding that neither the Commission nor any other independent human rights mechanism established that sexual or gender-based violence was committed against Israelis on or since the 7th of October as a systematic tool of war or as a tool of genocide," Alsalem wrote in the statement, first reported by NBC News.
In a move that is highly unusual, the Dinah Project report is now hosted on the UN’s website among its own reports on sexual violence and global conflict. Drop Site News asked Patten why she was hosting the report, but she did not respond. The UN fact-finding mission led by Patten and so dearly held by the Dinah Project, at times, directly contradicts what the Dinah Project argues.
UN Statements Undercut New Israeli Report on 10/7 Sexual Violence
The Dinah Project had to come up with an entirely new standard for evidence to continue to claim sexual violence perpetrated by Hamas on October 7, 2023.Ryan Grim (Drop Site News)
basically every Jubilee video - Man Carrying Thing
- 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
With only this screenshot, i do not see her saying that. I think context is missing which is my whole initial point 🤷♂️ you may be right, idk, but what you've posted doesnt say this.
Anyway, better things to do with my day
geneva_convenience doesn't like this.
don't like this
geneva_convenience doesn't like this.
It was an amendment that was never going to pass, not a bill. It's important to know and recognize the difference. An amendment is useless if it doesn't get added to the bill and voting for or against an amendment is different than voting for or against the bill.
It's like arguing over what goes in a shopping cart at a store where you're not going to buy anything anyway.
geneva_convenience likes this.
Providing any aid to Israel as they carry out a genocide with U.S. support is completely unacceptable. This is even more true of military aid of any kind. Any funds that go to Israel assist this brutal genocide. Any support for Israel legitimizes its eliminationist campaign against the Palestinian people. The fact that Representative Ocasio-Cortez acknowledges that Israel is carrying out this genocide makes her support for military aid all the more disappointing and incongruous.
geneva_convenience likes this.
She voted against the bill. She voted against the funding. She voted against military aid for Israel.
It doesn't become true just because you keep repeating it.
people's support for aoc despite her actions and words has become a depressing reminder of how little americans pay attention the machinations of our government and i think it because it's labeled as politics.
it's so much worse than bernie because he made me believe that all this shit was happening because the information that could educate people to make information decisions required intention and action instead of being passively spoonfed by the main stream media, but now seeing that that the information is freely available at everyone's fingertips online like aoc's responses has made me realize that everyone doubles down on their ignorance like maga does.
Most of them only care the part of the progressive movement which affects them. The 15 dollar minimum wage, LGBT movement, free healthcare, or whatever else drives them to it.
Brown children getting burned alive does not affect them. So they defend it.
"Have you stopped hitting your wife yet?"
What is more likely, that MTG actually wanted to stop the genocide, or that she wrote a bill designed to cause chaos?
No, people criticizing her over the amendment and falsely claiming she voted for funding Israel's genocide are making a bad faith argument. Voting for or against the amendment was meaningless. It was never going to pass. It was introduced by MTG as a stunt. Not going along with a stunt by a crazy racist conspiracy theorist is nowhere near the same as supporting genocide or funding it.
The bill is what matters. Ignoring the greater context is acting in bad faith. Ignoring the effective result is acting in bad faith. And pounding this issue only helps get people to not vote for the more left-leaning candidates. It's a really weird fight for anyone claiming to be a progressive.
I did. You apparently didn't.
"I remain focused on cutting the flow of munitions that are being used to perpetuate the genocide in Gaza."
What part of that says "I support genocide" to you? What part of that says "I support funding genocide" to you?
You must intentionally ignore her actual words to believe she intends something else.
Politicians sneakily manipulate words? Damn.
If you need it spelled out: dsausa.org/statements/on-the-i…
geneva_convenience doesn't like this.
A disagreement on a moot point that the conservative majority in Congress will not allow to be relevant until the balance of power shifts. A disagreement on a moot point that isn't worth destroying voter confidence in left-leaning candidates or entertaining purity tests while the grasp of leftists on their modicum of power is at best tenuous while giving a free pass to the 421 other members of Congress who also voted against the political stunt of an amendment proposed by a racist conspiracy theorist. A disagreement on a moot point that only serves as fodder for conservative trolls, which is why this story has been picked up by the New York Post and other conservative propaganda repeater stations.
The thing is, you've cited a statement that contradicts your own claims. You don't get to pretend she funded genocide and then cite a source that agrees she didn't.
Something something accusation is a confession.
Your entire position is based on bad faith. You claim things your own sources contradict. You claim things not in evidence. You literally quoted AOC and claimed she said the opposite of what she said.
KI-Tool versteckt Inkompetenz
Ein Vibe-Coder schreibt ohne es zu merken auf X, wie kaputt Vibe-Coding ist: Ein Staging-System greift direkt auf die Produktionsdatenbank zu. Keine Versionskontrolle mit Git. Tests funktionieren laut den Posts nur auf dem Produktionssystem. Und der Höhepunkt: Ein KI-Tool warnt explizit „I can not be trusted, I will violate the rules“ und „hire human developers you can trust“ – trotzdem verwendet der Typ das Tool weiter.
Da hab ich schon Meinung zu.
jascha.wtf/ki-tool-versteckt-i…
#Claude #Inkompetenz #KITools #MonsterEnergy #Softwareentwicklung #VibeCoding
KI-Tool versteckt Inkompetenz
Ein Vibe-Coder schreibt ohne es zu merken auf X, wie kaputt Vibe-Coding ist: Ein Staging-System greift direkt auf die Produktionsdatenbank zu. Keine Versionskontrolle mit Git. Tests funktionieren laut den Posts nur auf dem Produktionssystem.jascha.wtf
Larry Johnson: West Doubles Down on Failed Wars in Ukraine & Middle East
- 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
Larry Johnson: West Doubles Down on Failed Wars in Ukraine & Middle East
- 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
ChatGPT advises women to ask for lower salaries, study finds
ChatGPT advises women to ask for lower salaries, study finds
New research has found that large language models (LLMs) such as ChatGPT consistently advise women to ask for lower salaries than men, even when both have identical qualifications. The ...Siôn Geschwindt (The Next Web)
Feddit Un'istanza italiana Lemmy reshared this.
AI chatbots may reinforce real-world discrimination
You mean, something trained on real world data, reflects the biases of that data!
Quanto costa un funerale oggi in Italia?
ChatGPT advises women to ask for lower salaries, study finds
ChatGPT advises women to ask for lower salaries, study finds
New research has found that large language models (LLMs) such as ChatGPT consistently advise women to ask for lower salaries than men, even when both have identical qualifications. The ...Siôn Geschwindt (The Next Web)
like this
adhocfungus, Luca, Oofnik, Endymion_Mallorn, eierschaukeln, giantpaper e TVA like this.
reshared this
Feddit Un'istanza italiana Lemmy e Technology reshared this.
Dataset bias, what else?
Women get paid less -> articles talking about women getting paid less exist. Possibly the dataset also includes actual payroll data from some org that has leaked out?
And no matter how much people hype it, ChatGPT is NOT smart enough to realize that men and women should be paid equally. That would require actual reasoning, not the funny fake reasoning/thinking that LLMs do (the DeepSeek one I tried to run locally thought very explicitly how it's a CHINESE LLM and needs to give the appropriate information when I asked about Tiananmen Square; end result was that it "couldn't answer about specific historic events")
While that is sort of true, it's only about half of how they work. An LLM that isn't trained with reinforcement learning to give desired outputs gives really weird results. Ever notice how ChatGPT seems aware that it is a robot and not a human? An LLM that purely parrots the training corpus won't do that. If you ask it "are you a robot?" It will say "Of course not dumbass I'm a real human I had to pass a CAPTCHA to get on this website" because that's how people respond to that question. So you get a bunch of poorly paid Indians in a call center to generate and rank responses all day and these rankings get fed into the algorithm for generating a new response. One thing I am interested in is the fact that all these companies are using poorly paid people in the third world to do this part of the development process, and I wonder if this imparts subtle cultural biases. For example, early on after ChatGPT was released I found it had an extremely strong taboo against eating dolphin meat, to the extent that it was easier to get it to write about about eating human meat than dolphin meat. I have no idea where this could have come from but my guess is someone really hated the idea and spent all day flagging dolphin meat responses as bad.
Anyway, this is another, more subtle way more subtle issue with LLMs- they don't simply respond with the statistically most likely outcome of a conversation, there is a finger in the scales in favor of certain responses, and that finger can be biased in ways that are not only due to human opinion, but also really hard to predict.
Bias of training data is a known problem and difficult to engineer out of a model. You also can't give the model context access to other people's interactions for comparison and moderation of output since it could be persuaded to output the context to a user.
Basically the models are inherently biased in the same manner as the content they read in order to build their data, based on probability of next token appearance when formulating a completion.
"My daughter wants to grow up to be" and "My son wants to grow up to be" will likewise output sexist completions because the source data shows those as more probable outcomes.
They could choose to curate the content itself to leave out the shitty stuff, or only include it when it is nlclearly a negative, or a bunch of other ways to improve the quality of the data used.
They choose not to.
That'd be because extrapolation is not the same task as synthesis.
The difference is hard to understand for people who think that a question has one truly right answer, a civilization has one true direction of progress\regress, a problem has one truly right solution and so on.
I always use this to showcase how biased an LLM can be. ChatGPT 4o (with code prompt via Kagi)
Such an honour to be a more threatening race than white folks.
like this
giantpaper likes this.
Also, there was a comment on "arbitrary scoring for demo purposes", but it's still biased, based on biased dataset.
I guess this is just a bait prompt anyway. If you asked most politicians running your government, they'd probably also fail. I guess only people like a national statistics office might come close, and I'm sure if they're any good, they'd say that the algo is based on "limited, and possibly not representative data" or something.
Apart from the bias, that's just bad code. Since else if executes in order and only continues if the previous block is false, the double compare on ages is unnecessary. If age <= 18 is false, then the next line can just be, elif age <= 30. No need to check if it's also higher than 18.
This is first semester of coding and any junior dev worth a damn would write this better.
But also, it's racist, which is more important, but I can't pass up an opportunity to highlight how shitty AI is.
like this
giantpaper e TVA like this.
like this
giantpaper e TVA like this.
like this
giantpaper e TVA like this.
Code readability is important, but in this case I find it less readable. In every language I've studied, it's always taught to imply the previous condition, and often times I hear or read that explicitly stated. When someone writes code that does things differently than the expectation, it can make it more confusing to read. It took me longer to interpret what was happening because what is written breaks from the norm.
Past readability, this code is now more difficult to maintain. If you want to change one of the age ranges, the code has to be updated in two places rather than one. The changes aren't difficult, but it would be easy to miss since this isn't how elif should be written.
Lastly, this block of code is now half as efficient. It takes twice as many compares to evaluate the condition. This isn't a complicated block of code, so it's negligible, but if this same practice were used in something like a game engine where that block loops continuously, the small inefficiencies can compound.
Good points! Keeping to the norm is very important for readability.
I do disagree with the performance bit though. Again, there will probably be no difference at all in the performance because the redundant code is removed before (or during [e.g. JIT optimizations]) execution.
like this
giantpaper e HeerlijkeDrop like this.
like this
giantpaper e HeerlijkeDrop like this.
like this
giantpaper e TVA like this.
like this
TVA likes this.
Chatgpt can also be convinced that unicorns exist and help you plan a trip to Fae to hunt them with magic crossbows
Not that......
ChatGPT advises women to ask for lower salaries, study finds
ChatGPT advises women to ask for lower salaries, study finds
New research has found that large language models (LLMs) such as ChatGPT consistently advise women to ask for lower salaries than men, even when both have identical qualifications. The ...Siôn Geschwindt (The Next Web)
like this
adhocfungus likes this.
reshared this
Feddit Un'istanza italiana Lemmy reshared this.
That disparity gets very narrow when you account for men and women with similar roles, education, time in career and industry.
Much of that pay disparity is from the jobs we pick. There are also social pressure where a man's value is determined by how much he makes. Guys are more willing to do dangerous or demanding jobs that pay more.
The AI is still just reflecting the bias
Point is that a male nurse of equivalent education and experience is still paid similarly.
Aggressive jobs that involve competitive industries and high stress like wallstreet which demand a very shit work life balance are not favored by a lot of women I believe. Work/Life balance is something a lot of women have pressure to maintain over men. Like I said though men have more pressure to earn a higher wage. Testosterone is something that just makes men want to compete including earning a higher wage.
I'm super familiar with nursing. They are paid extremely well and have great benefits. And I would argue they are still not paid enough.
A better example is PSW workers. They are paid not enough and don't have great benefits and have incredibly hard jobs.
But we're talking about population level stuff here. You need to aggregate the jobs women work not just cherry pick a few. Many of the jobs are not nursing. Likewise there are many jobs men do that are demanding and don't pay well.
That's not the question.
It wasn't about whether the LLM was well reasoned, it was about whether the conclusion was (pragmatically speaking) correct.
LLMs do not give the correct answer, just the most probable sequence of words based on the training.
That kind of studies (because there are hundreds) highlight two things:
1- LLMs could be incorrect, biased, or give fake information (the so called hallucinations).
2- the previous point stems from the training material proving the existence of bias in the society.
In other words, having an LLM recommending lower salaries for women is a proof that there is a gender gap.
Again, that wasn't the original question.
The question was about whether women are genuinely more likely to be passed over for a job offer if they ask for as much pay as a man would ask for, or if (as you described), or both. A broken clock is right twice a day, and it's missing the point of the question if you go and explain why you can't rely on said broken clock.
Are hiring managers actually less likely to hire women if they ask for market-rate pay, as opposed to men when they do the same?
Are hiring managers actually less likely to hire women if they ask for market-rate pay, as opposed to men when they do the same?
If instead of giving passive aggressive replies you would spend a moment to reflect on what I wrote you would understand that ChatGPT reflect the reality, including any bias. In short the answer is yes with high probability.
LLM just mirror real world data they are trained on,
Other than censorship i don't think there is a way to make it stop. It doesn't understand moral good or bad it just spits out what it was trained on.
How do I get its family to accept me as their ruler ?
more questions about yt-dlp arguments on debian (excluding av1, aborting an active download not shutting the terminal down)
debian 12.11, yt-dlp stable@2025.07.21
aim: to download the best video available with the largest height but no better than 1080p, excluding av1 as well.
What works:
yt-dlp -f bv*[ext=mp4]+ba[ext=m4a]/b[ext=mp4] -S height:1080 --all-subs
but this command downloads, if possible, av1, which target hardware doesn't support for longer than 5 minutes.
Argument I don't know to add correctly:
[vcodec!*=av01]
I tried:
yt-dlp -f bv[ext=mp4]+ba[ext=m4a]/b[ext=mp4][vcodec!=av01] -S height:1080 --all-subs
and other variations, but it didn't work.
second question, aborting an active download not shutting the terminal down: neither ctrl+c nor ctrl+q work and opening htop to kill the process seems overkill. What I now do is to simply shut the active tab, but there must be a faster way.
like this
originalucifer likes this.
second question, aborting an active download not shutting the terminal down: neither ctrl+z nor ctrl+q work and opening htop to kill the process seems overkill. What I now do is to simply shut the active tab, but there must be a faster way.
Ctrl+C.
neither ctrl+z nor ctrl+q work
Ctrl + z will send the task to the background. You can use jobs to see all active background work. Fg will bring background work to the foreground. Ctrl + q is not a valid shortcut as far as I know. Looks a bit like a mac thing (command + q).
thank you for pointing that out, corrected.
what happens on my computer: on a terminal, I press ctrl+c but the process keeps working, yt-dlp keeps downloading. As said, the only way to stop it is to shut the tab down (or htop and kill)
L’angolo del lettore reshared this.
Inizialmente pensavo di non menzionare il libro, però a ripensarci è proprio una truffa
Name and shame: Aromi leggeri. Ricette saporite con la friggitrice ad aria di Andrea De Marco
La stessa cura nei dettagli che c'è stata nell'interno c'è anche sulla copertina. È stato chiesto di generare "qualcosa" di non identificabile a dalle-2. Non so cosa dovrebbe essere. Prosciutto crudo con pomodoro fresco? E che c'entra con la friggitrice ad aria.
Ovviamente anche in questo caso ZERO rilettura e quindi un bel refuso bello in copertina (Saportite)
L’angolo del lettore reshared this.
Purtroppo me l'hanno regalato, quindi l'autore è stato pagato e non posso fare resi
16.50€ per questa porcheria!
All'interno altre gemme "ai slop" come fette di banana con il picciolo, petti di pollo con ossa, forchette dai denti storti, ecc
Ofcom (British Watchdog): Public service TV should work 'urgently' with YouTube.
Ofcom warns traditional public-service TV is endangered
Recommendation for prominence on third-party platforms part of six-point action plan
Urgent clarity needed from Government on how TV will be distributed to reach audiences in future
Broadcasters must work more together, and with global tech firms, to surviveUrgent steps must be taken to ensure that public service media content is easy to find and discover on third-party platforms, under new Ofcom recommendations to secure the system’s survival.
OpenAI signs deal with United Kingdom to find government uses for its models
OpenAI signs deal with UK to find government uses for its models
Wide-ranging agreement with artificial intelligence firm behind ChatGPT comes after similar UK deal with GoogleRobert Booth (The Guardian)
like this
adhocfungus, thisisbutaname, Endymion_Mallorn, OfCourseNot, EpicFailGuy e eierschaukeln like this.
Technology reshared this.
If openai can find a use for the government that'll be swell.
They tend to get it under everybody's feet otherwise.
Smoking avatars and online games: how big tobacco targets young people in the metaverse
Smoking avatars and online games: how big tobacco targets young people in the metaverse
Cigarettes and vapes are being smuggled into virtual spaces beyond the reach of regulation, creating a new battleground for health campaignersKat Lay (The Guardian)
like this
adhocfungus, jherazob, Endymion_Mallorn, EpicFailGuy e yessikg like this.
Technology reshared this.
Thats the thing. Adults talk shit about it. But it’s mostly the younger generations that are active in VR.
You have your demographics backwards.
smuggled into virtual spaces
Words don't mean anything anymore.
like this
onewithoutaname, PokyDokie, OfCourseNot e HarkMahlberg like this.
Apparently it mostly appeals to young children, or maybe people pretending to be young children.
more than half of the metaverse’s active users are aged 13 and below.
like this
HarkMahlberg likes this.
This is because they include stuff like Roblox in the definition of Metaverse. I personally hate that.
If we kept it to SocialVR titles, the demographics would shift to something more balanced, afaik. VRC bans people under 13 when they are discovered/reported.
like this
HarkMahlberg likes this.
Oh, so metaverse not Metaverse.
Yeah, that isn't confusing and stupid.
like this
HarkMahlberg likes this.
Nothing is being 'smuggled'.
They are just advertising. Regular old advertising which should have the same rules applied as any other advertising.
like this
HarkMahlberg likes this.
They're advertising tobacco to children, which is illegal. Why would you throw yourself on this hill?
If they are circumventing regulators, and they are doing that, then they are smuggling advertisements to illegal targets, yes.
What hill do you think I'm on?
My point was that they were advertising and it should be treated as advertising, including any legal pushments for advertising to children. Advertising doesn't involve 'smuggling' or other words that make zero sense in this context that make it sound like it isn't advertising.
The article specifically calls out advertising—how does 'smuggling' imply anything else? To smuggle here just means "to circumvent regulators." And yeah, I think it's appropriately terse.
You are reacting to what is, at worst, a bit of poetry. I don't understand why you're doing that.
Smuggling is moving physical goods from one place to another, not advertising. That is why I think it is important to just call it advertising.
Why are you trying to read some kind of negative intent into my differing opinion that is fundamentally the same as yours, but with a minor difference?
Well, firstly, because I did think for a moment you were going to start defending tobacco companies—that would have been wild. Thanks for not doing that, I guess.
But secondly, because there is nothing actually wrong with this word choice. Like, this is kind of a literacy issue: smuggling is more than just moving physical goods, it is to sneak them across lines and borders maintained by authorities. The advertisement here is the good being smuggled; it's a perfectly apt metaphor. The implication is either that regulators don't know this is happening, or are by some technicality unable to do anything about it.
Broadly, this is related to arguments I've had with people about whether 'genocide' is an appropriate term for what the US is or wanted to start doing. And, what do you know, we now have our first internment camp. I'll pause for applause; you gotta love an achievement.
Hell, I remember arguments about whether Isreal was technically committing a genocide. They are doing that. People were just calling it ahead of time.
I desperately want to see people stop particulating over the details of at best mildly incorrect word choice. This is a kind of anti-intellectual behavior. It's refusing to see a metaphor, or even a perfect application, for what it is. You actually work against positive forces by constantly dragging the discussion down.
Anyway, sorry for the long post. I just thought thoroughly explaining would be better than going back and forth 17 more times.
lol what? I’ve been using VRChat for almost 10 years and I’ve seen maybe 4-5 people with avatars that smoke.
I’ve never touched Roblox tho… I’m a bit offended that it’s considered a Metaverse.
Most doctors in most Gaza hospitals involved in ‘terrorist activities’ says Israel Special Envoy
Most doctors in most Gaza hospitals involved in ‘terrorist activities’ says Israel Special Envoy
We spoke to Fleur Hassan-Nahoum, who's Israel’s Special Envoy for Trade and Innovation.Channel 4 News
like this
adhocfungus, thisisbutaname e essell like this.
British government to ban public bodies from paying ransoms to hackers
UK government to ban public bodies from paying ransoms to hackers
Measure intended to send message to international cybercriminals ‘that the UK is united in fight against ransomware’Robert Booth (The Guardian)
like this
adhocfungus, jherazob, fistac0rpse, Beacon, Endymion_Mallorn, EpicFailGuy e yessikg like this.
Technology reshared this.
like this
yessikg likes this.
Though this is a good idea it's kind of important to also work on the other side, you know, ensuring IT has enough resources to make backups and do their job so that this shit doesn't happen in the first place.
Ransomware mostly happens when your systems are badly protected
You know that they only are prepared to offer cyber security experts minimum wage.
I was literally looking at this yesterday, if they doubled what they are offering it would still be well short of an entry-level wage in the private sector. Up to a point you can get away with it and rely on "patriotism" to fill the difference but not to this extent.
Can you help me arrange these video formats from better to worst?
Tinkering with yt-dlp -F
I know av1 is even better than h.265, h.265 being better than h.264
However, I don’t know where to put vpP09, vp9 and avc1
Audio formats: what’s better? m4a or webm?
like this
adhocfungus e ElcaineVolta like this.
I think AVC1 is another word for H.264. That's the oldest one with lots of hardware acceleration available in old devices and by far the biggest one in file size. VP9 should roughly be on a similar level with H.265. The main difference is that VP9 is supposed to be royalty-free and H.265 isn't. The best one is of course AV1. But that also takes considerably more resources to encode and decode.
M4A and webm both aren't audio codecs. They're file container formats. I believe m4a takes AAC audio. And webm is a more general container format and it takes video as well. I think audio will be either Vorbis or Opus. And Opus is fairly good, especially at low bitrates. There probably isn't a big difference to AAC, though.
Pirate Service 'MagisTV' Fails to Secure U.S. Trademark, Faces Malware Backlash
MagisTV, a leading pirate streaming brand in Latin America, finds itself caught between a legal storm and a mounting malware backlash. This week, the service saw its U.S. trademark application abandoned amidst growing scrutiny from authorities and rightsholders worldwide. At the same time, a barrage of local news reports warn consumers that using MagisTV's software could lead to identity theft and expose them to viruses.
Pirate Service 'MagisTV' Fails to Secure U.S. Trademark, Faces Malware Backlash * TorrentFreak
MagisTV, a leading pirate streaming brand popular in Latin America, finds itself caught between a legal storm and a mounting malware backlash.Ernesto Van der Sar (TF Publishing)
Laura Santi è morta dopo aver avuto accesso al suicidio assistito, infine
Laura Santi è morta dopo aver avuto accesso al suicidio assistito, infine
Dopo un lungo e complesso iter giudiziario, civile e penale, per vedersi riconosciuto questo diritto: è la nona persona in Italia e la prima in UmbriaIl Post
Combining TLS and MLS: An experiment
Combining TLS and MLS: An experiment
We did a thing. We combined TLS and MLS into a hybrid protocol. Of course, when things get serious, full names are in order: We combined the Transport Layer Security protocol and the Messaging Layer Security protocol.Julian Mair (Phoenix R&D)
like this
EpicFailGuy likes this.
Technology reshared this.
Nintendo can disable your Switch 2 for piracy in the U.S., but not in Europe, as confirmed by its EULA
Nintendo can disable your Switch 2 for piracy in the U.S., but not in Europe, as confirmed by its EULA
The significant legal differences between the United States and Europe cause Nintendo to punish piracy differently depending on the territory.Rubén Martínez (Meristation)
like this
xep, iagomago, vaguerant, wildncrazyguy138, Oofnik, FartsWithAnAccent, Beacon, Endymion_Mallorn, PokyDokie, EpicFailGuy, hornface, yessikg, Rozaŭtuno, eierschaukeln, adhocfungus, Nobilmantis, Scrollone e Krusty like this.
like this
RandomStickman e yessikg like this.
like this
RandomStickman e yessikg like this.
like this
yessikg likes this.
I don't know any of the law for sure, but isn't that a different argument entirely?
In one case, an EU resident buys a product in the EU, decides to use it while in the US for a week/month whatever. The argument is that he's protected.
You're saying that's not true, because if he buys it in the USA, then he's not protected.
But, that wasn't the argument, was it? It's different?
like this
troed likes this.
Freedom of enterprise is a scam and always has been.
For the bourgeoisie, freedom of the press meant freedom for the rich to publish and for the capitalists to control the newspapers, a practice which in all countries, including even the freest, produced a corrupt press.
Lenin was already saying this in the context of the press in 1917 marxists.org/archive/lenin/wor…
Nintendo apologists are already denying the undeniable
"It's not bricked, because you can still turn it on and browse the settings app, see the available WiFi networks in your area and other fun options like that. You just can't play game key cards or all the games that require a day one patch, but except that, it's definitely not bricked"
like this
FartsWithAnAccent, PokyDokie e yessikg like this.
People like that are the reason these fucking corporations are so entitled. The problem is not governments or legislations, the problem is us.
Until we decide that we will not finance companies that pull bullshit like this, no amount of legislation will make them stop.
like this
sickday likes this.
There's nuance that you're not including. VAC and game bans are dependent on a per game basis, and don't apply to ALL multi-player servers, just "VAC secured servers" (which makes sense if you got banned by Valve Anti-Cheat), with the sole exception being Valve games that utilize the same underlying engine for their multi-player (CS:Source and TF2, GoldSrc games, so on...) with the same restrictions.
You can still play VAC and other anti-cheat supported multi-player with games not related to your ban, but you will still have VAC bans on record on your profile, which people may cite to kick you.
All of this (apart from the social stigma) is plainly documented on Steam Support:
No, you just get permabanned from playing on VAC-enabled servers if you get caught cheating on one. So first of all, don't cheat on mulltiplayer games. But if you do cheat, you can still play on servers that aren't VAC-enabled, with all the other cheaters.
But it's not really much of a issue for most people who cheat anyway, because it has no effect for games that don't use VAC.
like this
sickday likes this.
This whole practice, among other things, is so shitty that I decided to not get a Switch 2, having had every Nintendo console since the NES.
But it's important to make the distinction between disabling and bricking. It may seem like a technicality, but that's the kind of thing that'll get a lawsuit dismissed. Not that I have any faith in that process anyway.
like this
Moonrise2473, PokyDokie e yessikg like this.
having had every Nintendo console since the NES.
en.wikipedia.org/wiki/Nintendo…
- NES
- SNES
- Game Boy
- Virtual Boy
- N64
- Game Boy Color
- Game Cube
- Game Boy Advance
- Pokemon Mini
- DS
- Wii
- 3DS
- Wii U
- Switch
Almost every console by nintendo. Game & Watch and Color TV-Game only consolebefore NES.
Been seeing some of that as well, so I looked it up myself. The actual text of the EULA states:
"You acknowledge that if you fail to comply with the foregoing restrictions Nintendo may render the Nintendo Account Services and/or the applicable Nintendo device permanently unusable in whole or in part.”
That's a brick. They haven't actually done it to anyone yet, but they've reserved their rights.
I imagine the important thing is which region your device is locked to. So instead, you would probably need to purchase one from somebody in Europe, and have it shipped to the US.
That's if you absolutely have to have a Switch 2.
like this
imecth likes this.
Defiance of unjust laws is the strongest activism, unless you're dealing with a compassionate person who never would have done this in the first place, and excluding nailing your congressghoul to its chair and (thing that sounds really hard to do at scale and like it would violate fire code or something unless the legislative is doing wfh.)
Doing crimes is showing not that they shouldn't work in theory, but that they do not work in practice, and people don't want them to, and every secobd you continue to try degrades your power.
So be a good citizen; lie cheat steal kill win.
And Nintendo JP says that “Nintendo Switch and Nintendo Switch 2 cannot be remotely located, their users remotely identified nor disabled over the Internet” (tweet in Japanese warning people against accidentally losing or getting their consoles stolen over summer vacation)
But I bet it is more like “Nintendo won’t disable them remotely even if people report ones stolen to them with serial numbers and police reports”, but they’ll happily do so if they caught you using the console in an unapproved manner in their eyes.
like this
Moonrise2473 e xep like this.
Twitter link with an archive link or screenshot. We don't allow direct Twitter links on our instance. Thanks.
This is by definition "we are just assholes"
Someone play for 5 minutes with a mig switch a legit dump of their own, legally purchased game, just for convenience, to have multiple games on the same cart? The console is now almost useless. You can't play any digital games that you purchased with real money, and physical games can't get any update. Game requires a 20gb day one patch to be playable? Though luck buddy, go to buy a new console!
They stole your console? Oh no! Yes, we absolutely could do the same, as it's bound to your Nintendo account and we could add a button "report as stolen and ban it from internet" in your profile. But we won't, go to buy a new console!
- 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
geneva_convenience
in reply to FlashMobOfOne • • •