Dopo 11 giorni di negoziati a Ginevra, i delegati di 184 nazioni non sono riusciti a trovare un accordo su un trattato giuridicamente vincolante per affrontare la crisi globale della plastica.
I punti di disaccordo principali sono stati:
• Limitazione della produzione: profonda spaccatura tra chi chiedeva limiti vincolanti alla produzione di nuova plastica e chi si opponeva.
• Controlli chimici: stallo sull'imposizione di regole globali per le sostanze chimiche tossiche usate nella produzione.
• Finanziamento: nessun consenso su come finanziare l'attuazione del trattato, specialmente per i paesi in via di sviluppo.
La ministra francese per la transizione ecologica, Agnès Pannier-Runacher, si è detta "arrabbiatissima" per la mancanza di risultati tangibili, sottolineando che "la plastica uccide". Il delegato della Colombia ha accusato "un piccolo numero di stati" di aver bloccato l'accordo, in un apparente riferimento alle nazioni produttrici di petrolio che spingevano per focalizzarsi solo sul riciclo e non sulla riduzione della produzione.
I colloqui sono sospesi e dovrebbero riprendere in futuro, ma il fallimento ritarda una cruciale soluzione coordinata alla crisi.
Perché riguarda anche la moda?
Il poliestere è plastica. Questo trattato avrebbe avuto un impatto diretto sull'industria della moda, regolamentando le sostanze chimiche tossiche e la produzione della fibra sintetica più utilizzata dal fast fashion.
Each month, we create a post to keep you abreast of news and happenings regarding the server, discuss recent events, and to act as town square for the community.
🌟 Community Highlights 🌟
!abop@slrpnk.net by @oppression_abolisher (new community, help them get it started ♥️)
!smolweb@slrpnk.net - Dedicated to non-commercial, lean websites and its ideals. A concept becoming ever more appealing as the mainstream web tightens the noose with further enshittificiation.
!selfhosting@slrpnk.net - Goes hand-in-hand with smol web principles. Learn how to self-host your own websites and services for a more distributed and decentralized web!
🌏 World Carfree Day / In town, without my car! Events September 21-28 🚴
World Car-Free day is September 22nd. At SLRPNK we'd like to draw a little more attention to this holiday, and see it celebrated even more widely and internationally. Like all good holidays, it has its roots in civil disobedience.
In 1961, Jane Jacobs wrote The Death and Life of Great American Cities. She was one of the organizers of demonstrations that saved a popular public park in New York City from being turned into more road for cars. She was inspired by both the Garden City and City Beautiful movements before her, and her book and methods were internationally influential.
The Netherlands had seen the precursor to car-free days as a response to oil price shocks during the Suez Crisis, and anti-car sentiment has festered due to speeding vehicles in narrow city streets. Late one night in 1968, neighbors in Groningen took pickaxes and shovels to sections of their street to create intentional barriers that cars had to slow to navigate around. Despite official resistance, the civil disobedience movement to create "Woonerfs" or "Living Streets" spread, and in 1972 the first official Woonerf was constructed in Delft.
The Dutch lead the way in re-envisioning cities without cars through the 1970s, but the idea became internationally popular. People began organizing yearly car-free days, with the intention to explore other ways of organizing city life without the use of personal motor vehicles. Car-free days and automobile restraint in Urban planning goes hand in hand.
In one famous case in Jakarta, Indonesia, Car-free day is a weekly event. On Sunday mornings, several streets into the city are barred from use by traffic. They become popular paths for pedestrians and cyclists who come to the city for Sunday events. Initially it was planned to occur only 3 times a year in 2007, but as interest and infrastructure around the event grew, it became a weekly event only 5 years later. The streets used for this purpose have become more pedestrian-friendly as their use by pedestrians grew.
The most popular day for yearly Car-free days is September 22nd. In french-speaking countries, it is called "En ville, sans ma voiture" or "In town, without my car" - but the concept is the same. What ever the concept is called in your locality, we would like to boost it here as a Solarpunk holiday. Check out related communities for this event:
!urbanism@slrpnk.net
!utilitycycling@slrpnk.net
!publictransport@slrpnk.net
!TacticalUrbanism@slrpnk.net
📡 Technical updates from the servers 🧑💻
Not many updates from the technical department, but the consistent memory leak issue returned in Lemmy 0.19.12 that forces us to regularly restart the backend which always comes with a short 2-3 minute downtime (and if we don't catch it on time it sometimes crashes the database with causes issues with the XMPP auth integration).
There have been some nice improvements on our Movim instance though (OMEMO e2ee and image uploads should be more reliable now, and you can do full-text search in your chats), and we started experimenting with a XMPP server module that added Unified Push distributor support (so that you can use an XMPP app like Conversations to receive privacy preserving push notifications). Movim also added support for small group video calls (incl. screen-sharing), but we need to improve our STUN/TURN setup a bit for the connection establishment to be more reliable.
Last but least, some updates on the planned Piefed migration: The main blocker for re-using the bcrypt hashed passwords from the Lemmy database was resolved in Piefed which opens the path to start doing some testing on how to migrate accounts to the Piefed database. Don't expect immediate progress though, as time is limited to work on this right now. In addition the Hanubeki Lemmy theme we are using is also migrating to Piefed, so we will have a nice continuation of our color schemes.
💬 Open Discussion 💬
Now it’s your turn to share whatever you’d like down below; your thoughts, ideas, concerns, hopes, or anything related to the server. If you have a new community you’d like to shine a spotlight, shine away! If you’re a new user wanting to say hi, feel free to post an introduction 😀
SLRPNK Community Resources:
Community Wiki - Moderators, you can create your own Wiki here for your communities!
Movim Chat - Open to all members (use your SLRPNK login credentials)
UnifiedPush (UP) is an alternative for mobile push-notifications on Android that does not depend on the centralized Google FCM infrastructure for delivering notifications to your smart-phone.
En kiu lingvo oni parolu en nacia Esperanto-aranĝo? La demando estas same malnova kiel la organizita Esperanto-movado. Ofte oni plendas pri troa krokodilado en la nacia lingvo, sed en la ĵusa asembleo de Itala-Esperanto okazis male: oni oficiale malpermesis al la vicprezidanto de la asocio paroli en Esperanto.
@VM Evidente la ĉefa parto de la programeroj ja estis en Esperanto. (En la artikolo estas ligilo al la programo de la kongreso.) La malpermeso paroli en Esperanto okazis dum la asembleo de Itala Esperanto-Federacio, kiu estis nur malgranda parto de la kongreso.
Federal clean energy tax credits have been essential to the financing of wind and solar projects across the country, and a key part of states’ plans to transition to wind and solar power.
Following President Donald Trump’s moves to quickly phase out those credits, pending projects have a tight time frame to start construction before their eligibility expires. But states have long struggled to speed up permitting decisions, reduce regulatory hurdles and add new power to the grid. And the clock is running out.
“Every month counts,” said Patty O’Keefe, Midwest regional director at Vote Solar, a clean energy advocacy nonprofit. “[The tax credits] are the financial backbone of nearly every renewable energy project that’s currently in the pipeline.”
Charles Borges, the agency's chief data officer, alleged that more than 300 million Americans’ Social Security data was put at risk by DOGE officials who uploaded sensitive information to a cloud account not subject to oversight. His disclosure was submitted to the special counsel’s office on Tuesday.
“After reporting internally to management and externally to regulators, serious data and security and integrity concerns impacting our citizens’ most sensitive personal data, I have suffered exclusion, isolation, internal strife, and a culture of fear, creating a hostile work environment and making work conditions intolerable,” Borges added.
The Government Accountability Project, which is representing him in his whistleblower case, posted Borges' resignation letter on its website Friday evening. Borges declined to comment.
“He no longer felt that he could continue to work for the Social Security Administration in good conscience, given what he had witnessed,” his attorney Andrea Meza said in a statement. She added that Borges would continue to work with the proper oversight bodies on the matter.
Social Security’s chief data officer, Charles Borges, resigned after alleging officials mishandled sensitive data on 300M Americans, citing retaliation and a hostile workplace.
The New York Times, citing a statement from Mueller's family and people close to him, reported Sunday night that Mueller had been diagnosed with Parkinson's disease in the summer of 2021 and has had difficulty speaking.
A House committee investigating the DOJ's handling of the Jeffrey Epstein sex trafficking case has withdrawn a subpoena to former FBI Director Robert Mueller, citing the state of his health.
The World Travel & Tourism Council projected ahead of Memorial Day that the U.S. would be the only country among the 184 it studied where foreign visitor spending would fall in 2025. The finding was "a clear indicator that the global appeal of the U.S. is slipping," the global industry association said.
"The world's biggest travel and tourism economy is heading in the wrong direction," Julia Simpson, the council's president and CEO, said. "While other nations are rolling out the welcome mat, the U.S. government is putting up the 'closed' sign."
Travel research firm Tourism Economics, meanwhile, predicted this month that the U.S. would see 8.2% fewer international arrivals in 2025, an improvement from its earlier forecast of a 9.4% decline but well below the numbers of foreign visitors to the country before the COVID-19 pandemic.
A decline in foreign visitors traveling to the United States has stretched well into the summer. And tourism experts say the downward trend shows no immediate signs of reversing.
A plan circulating in the White House to develop the “Gaza Riviera” as a string of high-tech megacities has been dismissed as an “insane” attempt to provide cover for the large-scale ethnic cleansing of the Palestinian territory’s population.
Named the Gaza Reconstitution, Economic Acceleration and Transformation Trust – or GREAT – the proposal was reportedly developed by some of the same Israelis who created and set in motion the US- and Israeli-backed Gaza Humanitarian Foundation (GHF) with financial planning contributed by Boston Consulting Group.
Most controversially, the 38-page plan suggests what it calls “temporary relocation of all of Gaza’s more than 2 million population” – a proposal that would amount to ethnic cleansing, potentially a genocidal act.
Palestinians would be encouraged into “voluntary” departure to another country or into restricted, secure zones during reconstruction. Those who own land would be offered “a digital token” by the trust in exchange for rights to redevelop their property, to be used to finance a new life elsewhere. Those who stay would be housed in properties with a tiny footprint of 323 sq ft –minuscule even by the standards of many non-refugee camp homes in Gaza.
After his first arrest for sex crimes, Jeffrey Epstein tried to get into a new line of work: surveillance. In 2015, he partnered with former Israeli Prime Minister Ehud Barak to invest in a security tech startup called Reporty Homeland Security, now known as Carbyne. Leaked emails show that Epstein was using Barak to seek out opportunities in the surveillance industry and build connections with powerful figures around the globe, including American businessman Peter Thiel, the former director of Israeli signals intelligence, and two people in Russian President Vladimir Putin's circle.
After he was first caught sexually exploiting teenage girls, Epstein had pleaded guilty to soliciting prostitution in 2008; he served a little over a year in detention. Meanwhile, he invested his wealth in bizarre projects, including a ranch to breed women with his DNA and "efforts to identify a mysterious particle that might trigger the feeling that someone is watching you," according to The New York Times.
The leaked emails show that Epstein was also interested in more mundane means of spying on and manipulating people, which overlapped with the technologies governments often pursue. This interest crossed borders.
Barak's email inbox was quietly posted by Distributed Denial of Secrets, a website widely considered to be a successor to WikiLeaks, on a file-sharing platform for verified journalists and researchers in May 2025. The contents came from Handala, a hacker group named for a Palestinian cartoon character that has been leaking files taken from senior Israeli officials for several months.
Although the emails were posted without technical metadata or cryptographic signatures that would allow their authenticity to be verified, they include dozens of images, videos, voice recordings, and scanned documents from Barak and his friends and family that have never been published elsewhere. And they include information that was not publicly known at the time of the email leaks, including a reference to Epstein's birthday book.
The emails below, which have not been published elsewhere, paint a picture of Epstein as a man very eager to be at the nexus between private money and public surveillance. While they were hammering out the Reporty investment, Epstein invited Barak to come to a meeting with Thiel, cofounder of PayPal and the surveillance contractor Palantir, in May 2014. Although Barak couldn't make that meeting, Epstein insisted that Barak "spend real time with peter thiel [sic]" and offered to set up a dinner the following month.
Apple's alleged automation mandate spans all major product categories, including the iPhone, iPad, Mac, and Apple Watch. Apple now purportedly expects suppliers to fund their own automation upgrades rather than rely on Apple to finance or subsidize the necessary capital equipment. This policy change diverges from Apple's previous approach, where the company frequently invested in tooling and machinery for contract manufacturers to meet its specifications.
Thr reason of course is that thr soldiers are deployed to crack down on people with dark skin having rights or accumulating assets or political power, not any kind of actual criminal activities
That came in November, filed by Texas Attorney General Ken Paxton and 10 other Republican AGs, accusing three of the biggest asset managers on Wall Street—BlackRock, Vanguard and State Street—of running “an investment cartel” to depress the output of coal and boosting their revenues while pushing up energy costs for Americans. The Trump administration’s Department of Justice and Federal Trade Commission filed a supporting brief in May.
The overall pressure campaign aimed at what’s known as “ESG” is having an impact.
“Over the past several months, through this [lawsuit] and other things, letters from elected officials, state and federal, there has been a chilling effect of what investors are saying,” said Steven Maze Rothstein, chief program officer of Ceres, a nonprofit that advocates for more sustainable business practices and was among the earliest letter recipients. Still, “investors understand that Mother Nature doesn’t know who’s elected governor, attorney general, president.”
Earlier this month, a US District Court judge in Tyler, Texas, declined to dismiss the lawsuit against the three asset managers, though he did dismiss three of the 21 counts. The judge was not making a final decision in the case, only that there was enough evidence to go to trial.
::: spoiler 12 Key Findings 1. Age assurance can be done in Australia privately, efficiently and effectively: Age assurance can be done in Australia – our analysis of age assurance systems in the context of Australia demonstrates how they can be private, robust and effective. There is a plethora of choice available for providers of age-restricted goods, content, services, venues or spaces to select the most appropriate systems for their use case with reference to emerging international standards for age assurance. 2. No substantial technological limitations preventing its implementation to meet policy goals: Our evaluation did not reveal any substantial technological limitations that would prevent age assurance systems being used in response to age-related eligibility requirements established by policy makers. We identified careful, critical thinking by providers on the development and deployment of age assurance systems, considering efficacy, privacy, data and security concerns. Some systems were easier for initial implementation and use than others, but the systems of all technology providers with a technology readiness level (TRL) 7 or above were eventually capable of integration to a user journey. 3. Provider claims have been independently validated against the project’s evaluation criteria: We found that the practice statements provided by age assurance providers with a TRL of 7 or above fairly reflected the technological capabilities of their products, processes or services (to the extent applicable to the Trial’s evaluation criteria). Some of the practice statements provided have needed to be clarified or developed during the course of the Trial, but we observed that they offer a useful option for transparency of the capabilities of the available age assurance systems. Those with a TRL below 7 will need further analysis when their systems mature. 4. A wide range of approaches exist, but there is no one-size-fits-all solution for all contexts: We found a plethora of approaches that fit different use cases in different ways, but we did not find a single ubiquitous solution that would suit all use cases, nor did we find solutions that were guaranteed to be effective in all deployments. The range of possibilities across the Trial participants demonstrate a rich and rapidly evolving range of services which can be tailored and effective depending on each specified context of use. 5. We found a dynamic, innovative and evolving age assurance service sector: We found a vibrant, creative and innovative age assurance service sector with both technologically advanced and deployed solutions and a pipeline of new technologies transitioning from research to minimum viable product to testing and deployment stages indicating an evolving choice and future opportunities for developers. We found private-sector investment and opportunities for growth within the age assurance services sector. 6. We found robust, appropriate and secure data handling practices: We found robust understanding of and internal policy decisions regarding the handling of personal information by Trial participants. The privacy policies and practice statements collated for the Trial demonstrate a strong commitment to privacy by design principles, with consideration of what data was to be collected, stored, shared and then disposed of. Separating age assurance services from those of relying parties was useful as Trial participants providing age assurance services more clearly only used data for the necessary and consented purpose of providing an age assurance result. 7. Systems performed broadly consistently across demographic groups, including Indigenous populations: The systems under test performed broadly consistently across demographic groups assessed and despite an acknowledged deficit in training age analysis systems with data about Indigenous populations, we found no substantial difference in the outcomes for First Nations and Torres Strait Islander Peoples and other multi-cultural communities using the age assurance systems. We found some systems performed better than others, but overall variances across race did not deviate by more than recognised tolerances. 8. There is scope to enhance usability, risk management and system interoperability: We found opportunities for technological improvement including improving ease of use for the average person and enhancing the management of risk in age assurance systems. This could include through one-way blind access to verification of government documents, enabling connection to data holder services (like digital wallets) or improving the handling of a child’s digital footprint as examples. 9. Parental control tools can be effective but may constrain children’s digital participation and evolving autonomy: The Trial found that both parental control and consent systems can be done and can be effective, but they serve different purposes. Parental control systems are pre-configured and ongoing but may fail to adapt to the evolving capacities of children including potential risks to their digital privacy as they grow and mature, particularly through adolescence. Parental consent mechanisms prompt active engagement between children and their parents at key decision points, potentially supporting informed access. 10. Systems generally align with cybersecurity best practice, but vigilance is required: We found that the systems were generally secure and consistent with information security standards, with developers actively addressing known attack vectors including AI-generated spoofing and forgeries. However, the rapidly evolving threat environment means that these systems – while presently fairly robust – cannot be considered infallible. Ongoing monitoring and improvement will help maintain their effectiveness over time. Similarly, continued attention to privacy compliance will support long-term trust and accountability. 11. Unnecessary data retention may occur in apparent anticipation of future regulatory needs: We found some concerning evidence that in the absence of specific guidance, service providers were apparently over-anticipating the eventual needs of regulators about providing personal information for future investigations. Some providers were found to be building tools to enable regulators, law enforcement or Coroners to retrace the actions taken by individuals to verify their age which could lead to increased risk of privacy breaches due to unnecessary and disproportionate collection and retention of data. 12. Providers are aligning to emerging international standards around age assurance: The standards-based approach adopted by the Trial, including through the ISO/IEC 27566 Series [Note 1], the IEEE 2089.1 [Note 2] and the ISO/IEC 25000 [Note 3] series (the Product Quality Model) all provide a strong basis for the development of accreditation of conformity assessment and subsequent certification of individual age assurance providers in accordance with Australia’s standards and conformance infrastructure. :::
Australia’s government trial of age‑assurance tech to keep under‑16s off social media says social media age checks can be done, despite errors and privacy risks
::: spoiler 12 Key Findings 1. Age assurance can be done in Australia privately, efficiently and effectively: Age assurance can be done in Australia – our analysis of age assurance systems in the context of Australia demonstrates how they can be private, robust and effective. There is a plethora of choice available for providers of age-restricted goods, content, services, venues or spaces to select the most appropriate systems for their use case with reference to emerging international standards for age assurance. 2. No substantial technological limitations preventing its implementation to meet policy goals: Our evaluation did not reveal any substantial technological limitations that would prevent age assurance systems being used in response to age-related eligibility requirements established by policy makers. We identified careful, critical thinking by providers on the development and deployment of age assurance systems, considering efficacy, privacy, data and security concerns. Some systems were easier for initial implementation and use than others, but the systems of all technology providers with a technology readiness level (TRL) 7 or above were eventually capable of integration to a user journey. 3. Provider claims have been independently validated against the project’s evaluation criteria: We found that the practice statements provided by age assurance providers with a TRL of 7 or above fairly reflected the technological capabilities of their products, processes or services (to the extent applicable to the Trial’s evaluation criteria). Some of the practice statements provided have needed to be clarified or developed during the course of the Trial, but we observed that they offer a useful option for transparency of the capabilities of the available age assurance systems. Those with a TRL below 7 will need further analysis when their systems mature. 4. A wide range of approaches exist, but there is no one-size-fits-all solution for all contexts: We found a plethora of approaches that fit different use cases in different ways, but we did not find a single ubiquitous solution that would suit all use cases, nor did we find solutions that were guaranteed to be effective in all deployments. The range of possibilities across the Trial participants demonstrate a rich and rapidly evolving range of services which can be tailored and effective depending on each specified context of use. 5. We found a dynamic, innovative and evolving age assurance service sector: We found a vibrant, creative and innovative age assurance service sector with both technologically advanced and deployed solutions and a pipeline of new technologies transitioning from research to minimum viable product to testing and deployment stages indicating an evolving choice and future opportunities for developers. We found private-sector investment and opportunities for growth within the age assurance services sector. 6. We found robust, appropriate and secure data handling practices: We found robust understanding of and internal policy decisions regarding the handling of personal information by Trial participants. The privacy policies and practice statements collated for the Trial demonstrate a strong commitment to privacy by design principles, with consideration of what data was to be collected, stored, shared and then disposed of. Separating age assurance services from those of relying parties was useful as Trial participants providing age assurance services more clearly only used data for the necessary and consented purpose of providing an age assurance result. 7. Systems performed broadly consistently across demographic groups, including Indigenous populations: The systems under test performed broadly consistently across demographic groups assessed and despite an acknowledged deficit in training age analysis systems with data about Indigenous populations, we found no substantial difference in the outcomes for First Nations and Torres Strait Islander Peoples and other multi-cultural communities using the age assurance systems. We found some systems performed better than others, but overall variances across race did not deviate by more than recognised tolerances. 8. There is scope to enhance usability, risk management and system interoperability: We found opportunities for technological improvement including improving ease of use for the average person and enhancing the management of risk in age assurance systems. This could include through one-way blind access to verification of government documents, enabling connection to data holder services (like digital wallets) or improving the handling of a child’s digital footprint as examples. 9. Parental control tools can be effective but may constrain children’s digital participation and evolving autonomy: The Trial found that both parental control and consent systems can be done and can be effective, but they serve different purposes. Parental control systems are pre-configured and ongoing but may fail to adapt to the evolving capacities of children including potential risks to their digital privacy as they grow and mature, particularly through adolescence. Parental consent mechanisms prompt active engagement between children and their parents at key decision points, potentially supporting informed access. 10. Systems generally align with cybersecurity best practice, but vigilance is required: We found that the systems were generally secure and consistent with information security standards, with developers actively addressing known attack vectors including AI-generated spoofing and forgeries. However, the rapidly evolving threat environment means that these systems – while presently fairly robust – cannot be considered infallible. Ongoing monitoring and improvement will help maintain their effectiveness over time. Similarly, continued attention to privacy compliance will support long-term trust and accountability. 11. Unnecessary data retention may occur in apparent anticipation of future regulatory needs: We found some concerning evidence that in the absence of specific guidance, service providers were apparently over-anticipating the eventual needs of regulators about providing personal information for future investigations. Some providers were found to be building tools to enable regulators, law enforcement or Coroners to retrace the actions taken by individuals to verify their age which could lead to increased risk of privacy breaches due to unnecessary and disproportionate collection and retention of data. 12. Providers are aligning to emerging international standards around age assurance: The standards-based approach adopted by the Trial, including through the ISO/IEC 27566 Series [Note 1], the IEEE 2089.1 [Note 2] and the ISO/IEC 25000 [Note 3] series (the Product Quality Model) all provide a strong basis for the development of accreditation of conformity assessment and subsequent certification of individual age assurance providers in accordance with Australia’s standards and conformance infrastructure. :::
This document presents the official report of the Age Assurance Technology Trial, offering a comprehensive overview of its findings, methodologies and key observations.
so much fluff in their findings. makes me think, they didnt research shit. this is gonna be terribly implemented.
their attempt wont be the end of the free web. but people will leave because they dont want their data breached by yet another bad actor. its just making the web more dangerous instead of safer.
no one has the tech and security to handle this. its waaaay too early. the breaches are going to substantial and continuous.
They've been trying to do it for ages, and they're forever a tad closer. The software stack is where the challenge lies, not the hardware. Still today, you can buy AMD hardware that on paper is better than Nvidia's, yet you can't squeeze out similar performances matching shittier Nvidia cards. Even still, rocm, the 'cuda' for AMD, can't even compete with even vulkan (an open source agnostic backend). So I doubt china will deliver that fast.
Software like wgpu makes it much easier to close the gap between various GPUs. New compute languages that are backend-agnostic are appearing, in the same vein as taichi-lang, that make it significantly easier to make high-performance gpu kernels deployable anywhere.
The compute groundwork for crossplatform tensor calculations is already here. Inference is already doable on any device. Training is not far behind. As a side-effect of this, processing on the GPU in every capacity, like physics, novel rendering techniques, or whatever else the imagination can muster, is now within grasp of "average" programmers.
If you have always been intimidated by GPU programming, I urge you to take another look now. The landscape is radically different. The software moat everyone talks about with NVIDIA is smoke-and-mirrors. Cuda is old news, though I am speaking to the actual code landscape here, not the common mental consensus.
What we lack now is cheap video cards that have high memory. I believe the current cards are overpriced by about 10 - 100x what they should be, because this profit situation is extremely temporary. Just as pens were once thousands of dollars, these compute devices will be collapsing in price.
I welcome China building cheaper video cards. Hopefully we will all benefit from it before any robot wars break out.
Weird article IMO??? Forget the startups, unless they are backed by very major players they don't stand a chance. They have their competences, but I seriously doubt competing with Nvidia on making AI chips is among them. The players to watch are Huawei, Baidu, Tencent and the likes. Who have already been working on this for a while, and have actual working and useful products.
While Huawei is the leader, Chinese companies don’t want to rely entirely on the company.
I don't understand how that statement is supposed to make much sense? When Chinese companies were happy using Nvidia and being dependent on Nvidia. Why wouldn't they be equally happy using Huawei if it's the best option after the government has forbidden them from using Nvidia? It may be true, but there is zero explanation why it is.
To the ones that think China can just use AMD instead, they really can't, AMD is under the same restrictions Nvidia is, and AFAIK AMD has not designed a chip to sell to China within those restrictions.
::: spoiler Key Findings 1. Geopolitical Events Trigger Unprecedented DDoS Campaigns Expand: Major political events drove increased DDoS activity, evidenced by attack count spikes that coincide with these occurrences. These events saw hacktivist groups launching up to double the normal number of attacks in short timeframes. 2. Botnet-Driven Attacks Dominate with Increased Sophistication Expand: Botnet-driven attacks are getting longer, more frequent, and are employing multiple attack vectors to avoid mitigation. They are targeting known vulnerabilities in IoT devices, servers, routers, and more. 3. NoName057(16) Maintains Dominance Among Familiar Threat Actors: Well-known hacktivist and attack groups, such as NoName057(16), are launching more attacks across the globe while leveraging several attack vectors. 4. New Threat Actors Emerge with DDoS-as-a-Service Capabilities: Emerging attack groups like DieNet and Keymous+ are leveraging DDoS-for-hire infrastructure to launch DDoS-as-a-service campaigns, lowering the barrier to entry and expanding the threat landscape. 5. Global DDOS Attack Volume High with Regional Variations: With more than 8 million recorded attacks globally in the first half of 2025, DDoS attack volume remains massive. The attacks also show sustained intensity, reaching speeds of 3.12 Tbps and 1.5 Gpps. :::
DDoS attacks are no longer just a nuisance, they’re a weapon of geopolitical influence. In the first half of 2025 alone, more than 8 million attacks were recorded globally, with threat actors leveraging AI, botnets, and DDoS-for-hire services to launch increasingly sophisticated and sustained campaigns.
DDoS Dominate the Digital Battlefield: AI integration, persistent hacktivist campaigns, and nation-state actors weaponize DDoS attacks, creating unprecedented risks for organizations globally
::: spoiler Key Findings 1. Geopolitical Events Trigger Unprecedented DDoS Campaigns Expand: Major political events drove increased DDoS activity, evidenced by attack count spikes that coincide with these occurrences. These events saw hacktivist groups launching up to double the normal number of attacks in short timeframes. 2. Botnet-Driven Attacks Dominate with Increased Sophistication Expand: Botnet-driven attacks are getting longer, more frequent, and are employing multiple attack vectors to avoid mitigation. They are targeting known vulnerabilities in IoT devices, servers, routers, and more. 3. NoName057(16) Maintains Dominance Among Familiar Threat Actors: Well-known hacktivist and attack groups, such as NoName057(16), are launching more attacks across the globe while leveraging several attack vectors. 4. New Threat Actors Emerge with DDoS-as-a-Service Capabilities: Emerging attack groups like DieNet and Keymous+ are leveraging DDoS-for-hire infrastructure to launch DDoS-as-a-service campaigns, lowering the barrier to entry and expanding the threat landscape. 5. Global DDOS Attack Volume High with Regional Variations: With more than 8 million recorded attacks globally in the first half of 2025, DDoS attack volume remains massive. The attacks also show sustained intensity, reaching speeds of 3.12 Tbps and 1.5 Gpps. :::
DDoS attacks are no longer just a nuisance, they’re a weapon of geopolitical influence. In the first half of 2025 alone, more than 8 million attacks were recorded globally, with threat actors leveraging AI, botnets, and DDoS-for-hire services to launch increasingly sophisticated and sustained campaigns.
On August 28 and 29, California Governor Gavin Newsom unveiled two sweeping initiatives that together mark a sharp rightward turn in state policy and expose the Democratic Party’s deepening complicity in the destruction of democratic rights. As he portrays himself as a bulwark against President Trump, Newsom is in fact laying the foundation for a massive expansion of state power against the working class and the poor.
Under the guise of public safety and compassion, the Democratic governor has placed the California Highway Patrol (CHP) at the center of two major new enforcement regimes: a statewide “crime suppression” expansion and a “homeless encampment clearance” task force.
These measures are being marketed as alternatives to Trump’s deployments of federal forces into major U.S. cities, but in substance, they mirror their basic functions. Far from opposing the authoritarian measures emanating from Washington, Newsom’s actions mimic them, signaling a growing alignment between the Democratic Party and the Trump administration on the fundamental issue: the use of state repression to deal with the social crisis created by capitalism.
In 2023, the then-Conservative government awarded Palantir with a £330m contract to “build a new NHS data platform”. The move proved controversial because Palantir is a US spy tech firm and its co-founder Peter Thiel is a prominent Trump donor with links to convicted sex offender Jeffrey Epstein. Now, a trove of leaked emails have shone further light on Thiel’s connections.
Journalists have reported on Thiel’s links to Epstein for some time. In 2019, journalist Whitney Webb reported on the links between Epstein, Israeli tech company Carbyne, and the American intelligence apparatus. In the piece, she reported:
Not long after Epstein’s arrest, and his relationships and finances came under scrutiny, it was revealed that the Israeli company Carbyne911 had received substantial funding from Jeffrey Epstein as well as Epstein’s close associate and former Prime Minister of Israel Ehud Barak, and Silicon Valley venture capitalist and prominent Trump backer Peter Thiel.
Webb added:
Another funder of Carbyne, Peter Thiel, has his own company that, like Carbyne, is set to profit from the Trump administration’s proposed hi-tech solutions to mass shootings. Indeed, after the recent shooting in El Paso, Texas, President Trump — who received political donations from and has been advised by Thiel following his election — asked tech companies to “detect mass shooters before they strike,” a service already perfected by Thiel’s company Palantir, which has developed “pre-crime software” already in use throughout the country. Palantir is also a contractor for the U.S. intelligence community and also has a branch based in Israel.
The latest revelations around Thiel come from the leaked emails of the aforementioned Ehud Barak (former Israeli prime minister). Writing in Reason, Matthew Petti reported:
Epstein invited Barak to come to a meeting with Thiel, cofounder of PayPal and the surveillance contractor Palantir, in May 2014. Although Barak couldn’t make that meeting, Epstein insisted that Barak “spend real time with peter thiel [sic]” and offered to set up a dinner the following month.Epstein
Barak wrote to a different business associate a few days later, without mentioning Epstein’s role, that he and Thiel would have a “first date” and “probably spend it talking just geopolitics” with an unnamed third person. In that email, Barak added that he had met Thiel once before in Davos, Switzerland, but speculated that Thiel “probably doesn’t even recall it.”
Brace Belden of the TrueAnon podcast added that Barak suspected Thiel was “under some drug impact” when they met, according to the emails:
Another funder of Carbyne, Peter Thiel, has his own company that, like Carbyne, is set to profit from the Trump administration’s proposed hi-tech solutions to mass shootings.
This willingness to let billionaires offer for-profit tech solutions to social problems is a sickness, and this is the clearest example yet.
Makes sense. Sex trafficking is sex trafficking. Remember when Lindsey Graham made the sudden switch from vehemently against the Orange pedophile/rapist to best buds? Blackmail makes real sense.
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.
Thiel shared his views on the Ten Commandments and identified which ones he believes are the most significant.
"The Ten Commandments, the two most important are the first and last on the list. The first commandment is, you should worship God," Thiel reportedly told attendees. "The tenth commandment is, you should not covet the things that belong to your neighbor. In some ways, the first commandment is to look up, and the tenth commandment is you do not look around. And if you're too much focused horizontally on all the people around you, that's sort of the bad version you get caught up in."
He's such a hypocrite!
The greedy guy who owns a company that offers mass surveillance systems says that not looking around at your neighbors is one of the best virtues. LoL
The liability is obviously not high enough if keeping and sharing this data works out positively. Even if the parties involved weren't obviously evil, the risk to the people is too high for them to accept the risk with only trivial, speculative improvements.
Yeah I'm not surprised about the Thiel/Epstein thing, I'm more surprised that the NHS is considering any sort of private health data handling by PALANTIR of all companies.
Somebody is going to be getting a Non-executive board membership or millionaire consulting "gig" from one of Tiel's companies or one from one of his friends...
By this point in time in Britain that kind of I thing has been going on long enough to almost be tradition.
NHS England has tens of millions of patients. Most patient records are undigitised and must be couriered when patients move to a new area or when specialist care is required.
It’s also a huge, publicly funded service so big data can be crunched to improve outcomes and efficiencies.
Palantir wanted a big system to use as the testbed for its move into healthcare data systems and offered a low bid.
Unless I am presented with contradictory evidence (I am not accepting feedback) I will assume nobody over there is willing to program a machine of any type out of fear that their balls will be cut off by the government like daddy Turing’s were.
Edit: (to be clear, I’m just taking the piss. Please for the love of dog, do not touch Palantir with even the tip of any of your extremities.)
This isn't even the first time Thiel has been linked to Epstein, but we never really hear about it for some reason. This isn't even the first time he's been linked this summer.
Ever since that gawker lawsuit, it kinda seems like bad press about Peter Thiel gets buried almost as soon as it's published.
as a rule I don't subscribe to theories related to the blood libel myth but if there was a dude to try injecting children's's blood into his blood to prolong his own life, it's probably this psycho. the dude needs a straight jacket and glands removed. not government contracts.
In 2016, various media outlets reported that Mr Thiel had links to the radical life extension startup Ambrosia, with Gawker claiming he "spends $40,00 per quarter to get an infusion of blood from an 18-year-old based on research conducted at Stanford on extending the lives of mice."
These reports cited his investment portfolio, together with a 2009 essay that laid out his philosophical and political beliefs. In it, he wrote that he stood against "the ideology of the inevitability of the death of every individual."
However the blood transfusion claims were never verified and Gawker shut down shortly after following an unrelated lawsuit partly funded by Mr Thiel.
The comedy show “Silicon Valley” was a documentary. They just changed the names of prominent figures and literally nothing else. That’s why it was so funny.
I don't know which all aircraft were involved at different points, but there was absolutely a shoot-down by an F-22.
I don't know anything about this website but it was one of Wikipedia's sources. I remember seeing the pictures/videos with the F-22 clearly visible, plus the noteworthy stats like being the first f-22 kill, being the first engagement over US territory since WW2, and that it was probably the highest air to air kill.
They provide police state technology currently used by ICE to target immigrants and several US cities police departments use their tech for "predictive policing." Not to mention they are actively and enthusiastically testing predictive generative AI to plan war crimes and hallucinate targets in the Gaza genocide. Technology they will bring home as "battle tested."
I assume y’all on the other side of the pond are just learning about Palantir for the first time and that’s why you’re looking for bombshell headlines. I can assure you as someone from Silicon Valley that there is literally no single headline that can possibly convey the pure evil that is Peter Thiel. He literally believes that it is morally and, from his perspective, objectively correct that a few pre-selected technocrat elites (his words) should rule city-states that “compete” with each other. He is currently miserably failing at building a libertarian “eutopia” (his words) out in the desert to the east of San Francisco. He promises it would be done by this year. It hasn’t even started.
He founded PayPal and got mad when he immediately learned that banks exist for a reason. He was such a piece of shit that Elon Musk couldn’t work with him.
Being a "thinker" isn't exactly a high bar. A fish can do it (depending on the definition). I mean I suppose I can think, but I wouldn't brag about it.
if you have nothing worth bragging about, resorting to basic functions of sapient life for clout is the best you can do. I'm a thinker too. I'm also a breather, an eater and plenty moee things that aren't special.
I've got some actual achievements too, such as being a former piece of shit, so I've actually got a leg up on him!
The neoreactionary movement (a.k.a. neoreaction, NRx, the Dark Enlightenment) is a loosely-defined cluster of Internet-based political thinkers who wish to return society to forms of government older than liberal democracy.
It is an ideal belief system for soi-disant libertarians who realize that, regardless of the principles of freedom being the ostensible foundation of their ideology, they are unwilling to tolerate others' enjoyment of freedom.
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.
The first time? Hardly, for anyone who can read and find proper news sources. I've been telling people for years that the most dangerous people in the fascist middle northern america are not in the white house, but in the silicon valley - and that the most evil motherfucker is Peter Thiel.
I can assure you as someone from Silicon Valley that there is literally no single headline that can possibly convey the pure evil that is Peter Thiel. He literally believes that ...
Dude, the US has been lagging behind the UK for the last decade. We voted for Brexit because of Cambridge Analytica, then you voted Trump for the first time because of Cambridge Analytica. We voted for Boris in 2019, you got Trump again in 2024. We're probably going to vote in Nigel Farage next and then the US will be like "Hold my beer!"
Palantir is just another symptom of this wider problem.
The recent fury over news that data on Facebook members was used by the Trump campaign exposes a massive double standard on the part of those now raising holy hell over Facebook.
uhm idk who's not familiar with Thiel but the blueprint of his "utopia" - "The Sovereign Individual" was literally co-written by no other than William Rees-Mogg a conservative house of lords cross-bencher who's also the father of two other conservative MPs that have served in the UK Parliament.
the 2014 reprint of the book has a foreword by Thiel, where he says that it's "the most influential book he had read"
so as usual the theoretical frameworks have been drafted in the heart of empire and implemented in the "land of the free"
Well, Elon wouldn't work with him because Elon had a shitty idea that Thiel said was shitty and wouldn't let Elon do it. Elon was forced to settle for becoming filthy rich.
That project is where the name "X" originally comes from.
Lol what the fuck? I assume you are taking about California Forever but desert is a weird word to describe the Sacramento River delta. It is one of the largest estuaries in north america.
It does seem like the muttering, scrawny weirdo would be the first to get put out on their arse post-apocalypse by... literally anyone currently in their employ.
Does he put forward his case that he is the anti christ? As he sure comes across as someone who would want that. He has always felt like the sort of motherfucker who would read the Stand and identify with Randall Flagg
I setup a Mastodon relay - anyone want to help me test by adding it to their instance? Would help me know if the "Recent jobs" stat is working (I think it requires 2 instances at minimum to show jobs) and if adding to instances (outside of my own) is working properly and how traffic looks.
The last I read, de minimis still applied. I didn't know until now that was done with.
As an avid collector of vinyl records: FUCK! I've got no problem sending $50 to a European artist who's selling a limited run of records out of their living room. Hell, if it's an artist I really like, I'll spend $70. I'm not about to spend $70 and the artist get half of it.
Spending ludicrous amounts of cash of 12-inch pieces of plastic is totally fine with me, but I want my money going to the artist who's making the music I love, not a government I voted against.
No more Lego pick a brick! Way more money for temu bullshit, if they even still ship here! Etc
This is where American consumerism will really start to feel the squeeze. The prices of all that stuff had gone up because of tariffs related to manufacturing costs but now direct tariffs on shipments will either block it or cost consumers.
Not only that, but no more pick-a-brick in Canada too! Why does my nephew have to suffer because Trump is a dink?
(We've tried maybe suggesting he not leave tiny Lego walkie talkies and lightsabers around for the dog to process - like a lab test but not a Labrador - but that's proven unworkable)
Bulk shipping requires storage and logistics infrastructure. Either way Temu and the likes will also be hit by increased costs, though I can't say if the impact is comparable.
There are many good things about this, American consumerism is out of control
We discuss climate change and how “companies are the worst offenders” but what drives those companies? American consumerism
Importing fast fashion, cheap plastic bullshit, other nonsense in plastic packaging, etc (much of it produced in countries that still utilize very dirty fossil fuel chains) ultimately funds and drives significant demand to keep it going and expand.
Also puts huge demand on fuel for international shipping of dumb bullshit.
Next thing to do would be to further reduce fuel demands by limiting air travel and consumer fuel usage but Trump isn’t going to invest in public transport, obviously. This is only a byproduct of his idiocy. After that would be to address concrete and other building material demand/suburban sprawl. Although the time to do this was 20 years ago
I’ll spend $70. I’m not about to spend $70 and the artist get half of it.
You're spending too much time listening to Trump/media apologists for Trump. The America hating foreigners don't pay the tariffs. If the artist you are giving $70 to, is shipping from a country with 25% tariffs, then US Customs, if foreign Post office did not collect US tariffs (all of them are refusing), or disagrees with the value/amount collected, then they will add anywhere between 25% of the missing value or $200 (tariffs on $800 value as penalty for not complying with US law), and YOU NEED TO PAY to collect the package.
Artist got the $70. They won't ship if they have to pay the tariffs. You pay the tariffs if they don't.
If the person in the US is only willing to pay 70, then the artist will get less than 70.
Sure, the American pays the tariff. That's not the point. The point is that this person wants to spend X total and wants most of that X to end up with the artist. And that doesn't happen if they have to spend half of X to pay for the tariff.
The point of saying "the American pays the tariff" isn't to say that the seller makes the same amount. It's to emphasize that the seller won't absorb that cost, which is the lie Trump is selling.
Considering the typical ebay listing, price is payment to be received. Shipping and tarrifs are extra to "weird jurisdictions". I get that OP is willing to offer artist/collector $70, but the seller wants that $70, and shipping somewhere human is preferred. The chaos of US policy is that they are telling all foreign countries to collect tariffs for them, which is not at all how customs/tariffs have ever worked globally ever before. US's fucking job to collect tariffs. No one is complying with absurdity.
The last I read, de minimis still applied. I didn't know until now that was done with.
You can blame companies like Temu for putting a spotlight on de minimis. Their entire business model was built around exploiting de minimis to never pay any taxes. Rather than importing a single shipping container valued above the de minimis amount, they list it as like 10000 individual items, each under the de minimis limit.
It was overwhelming port authorities who didn’t have the manpower to handle that much paperwork for what should have been listed as a single shipment. The tariffs originally didn’t touch de minimis, but then the feds noticed that companies were essentially evading tariffs by only shipping low value items.
Large corpos might set up fulfilment centers in USA and do bulk import. Small companies will just stop export to USA. Setting up manufacturing in USA is unlikely with all the volatility.
The Delhi High Court ordered the blocking of Sci-Hub, Sci-Net, and LibGen in India on August 19, 2025, following a copyright infringement case brought by academic publishers Elsevier, Wiley, and the American Chemical Society[^5][^7].
The court found that Alexandra Elbakyan, Sci-Hub's founder, violated her December 2020 undertaking not to upload new copyrighted content by making post-2022 articles available through both Sci-Hub and a new platform called Sci-Net[^7]. While Elbakyan claimed this was due to technical errors and argued Sci-Net was a separate project, the court rejected these arguments[^7].
The ruling requires India's Department of Telecommunications and Ministry of Electronics and Information Technology to issue blocking orders within 72 hours, with Internet Service Providers required to implement the blocks within 24 hours[^7].
This case marks the first time Sci-Hub and LibGen faced legal action in a developing country[^2]. Earlier intervention attempts by Indian scientists and researchers had argued these platforms were "the only access to educational and research materials" for many academics in India[^2], with social science researchers specifically highlighting the "detrimental effect" blocking would have on research in India[^9].
A group of social science researchers have filed an intervention application, with legal support from IFF, highlighting the adverse impact any decision to block LibGen and SciHub will have on them.
The natural world is built for change. Seasons shift. Rivers rise and fall. The climate gradually warms and cools again. Animals migrate, adapt, and evolve in response to these rhythms. This is how Earth has always worked – and how it’s supposed to work.
The pine forests of the Western U.S. offer a perfect example. For thousands of years, ponderosa and lodgepole pines evolved with periodic wildfires that swept through every decade or two. These fires weren’t disasters – they were essential.
Lodgepole pines actually depend on fire to reproduce. Their resinous cones only open in intense heat, releasing seeds onto the ash bed below. Ponderosa pines developed thick, fire-resistant bark to survive the low-intensity ground fires that cleared out undergrowth. These frequent, cool burns created open forests with widely spaced mature trees, healthy and highly productive ecosystems that provided clean water, timber, and wildlife habitat.
So the problem today isn’t change. It’s the speed of change.
Changes that used to take centuries or millennia are now unfolding in a matter of years. Levels of climate-warming carbon dioxide in the atmosphere have risen to well above 400 parts per million, a concentration that last occurred about 15 million years ago.
But even more concerning is the rate of change: By burning fossil fuels, we are emitting carbon dioxide into the atmosphere 30 times faster than at any point in the last 100 million years. That’s like putting nature’s slow-moving film on fast-forward – only the device is overheating as a result.
Juliana Moreira ed Edoardo Stoppa, ospiti di Caterina Balivo a La Volta Buona, hanno ripercorso i 17 anni della loro storia d'amore, mostrando il lato più autentico e ironico della loro coppia... leggi
How OnlyFans Piracy Is Ruining the Internet for Everyone
The internet is becoming harder to use because of unintended consequences in the battle between adult content creators who are trying to protect their livelihoods and the people who pirate their content.
Porn piracy, like all forms of content piracy, has existed for as long as the internet. But as more individual creators who make their living on services like OnlyFans, many of them have hired companies to send Digital Millennium Copyright Act takedown notices against companies that steal their content. As some of those services turn to automation in order to handle the workload, completely unrelated content is getting flagged as violating their copyrights and is being deindexed from Google search. The process exposes bigger problems with how copyright violations are handled on the internet, with automated systems filing takedown requests that are reviewed by other automated systems, leading to unintended consequences.
These errors show another way in which automation without human review is making the internet as we know it increasingly unusable. They also highlight the untenable piracy problem for adult content creators, who have little recourse to stop their paid content from being redistributed all over the internet.
I first noticed how bad some of these DMCA takedown requests are because one of them targeted 404 Media. I was searching Google for an article Sam wrote about Instagram’s AI therapists. I Googled “AI therapists 404 Media,” and was surprised it didn’t pop up because I knew we had covered the subject. Then I saw a note from Google at the bottom of the page noting Google had removed some search results “In response to multiple complaints we received under the US Digital Millennium Copyright Act.”
The notice linked to the Lumen Database, which keeps a record of DMCA complaints, who filed them, and for what. According to the Lumen Database, the complaint was filed by a company called Takedowns AI on behalf of content creator Marie Temara. Takedowns AI is one of many companies that help content creators, especially adult content creators, to scan the internet for images and videos they posted behind paywalls on platforms like OnlyFans and posted elsewhere for free. These companies also file DMCA takedown requests and navigate the copyright systems of big platforms like YouTube, Instagram, and Reddit. One of the most effective ways of preventing people from finding this pirated content is sending DMCA takedown requests to Google asking the search engine to delist results to sites that share it. As its name implies, Takedowns AI heavily relies on automation to do this work.
The complaint that impacted 404 Media included a list of 68 links to different websites that allegedly violated Temara’s copyright on content she posted to Instagram, OnlyFans, and other platforms. This was the allegedly offending link on 404 Media, which is a collage Sam made for her AI Facebook therapists story.
The collage includes Meta's AI-generated profile pictures of three of these AI therapists. The story itself has nothing to do with Temara, and the profile pictures look nothing like her. In fact, it would be hard for anyone to claim copyright for that image because in 2023 a U.S. court ruled that AI-generated art can’t be copyrighted.
I went through every other link in the same complaint and couldn’t find even one link that looked like it violated Temara’s copyrights. There were images from Grand Theft Auto V, famous baseball players, robots, stock images of people at theme parks, and movie posters, none of which looked remotely similar to Temara. In addition to 404 Media’s article about AI therapists, some of the pages that Google removed from search results due to this complaint included tech site wccftech.com, horror movie site bloody-disgusting.com, and rugby and wrestling sites.
These links are also just part of one complaint out of hundreds that Takedowns AI files every day. I looked through dozens of complaints to Google filed by Takedowns AI that were archived by the Lumen Database. The vast majority of them appear to be legitimate, but I did find other egregious mistakes. One of the worst mistakes I saw was a takedown request filed on behalf of a creator who goes by “honeyybee” against an article about actual honey bees on the University of Missouri’s website. The takedown request clearly targeted the article and caused Google to remove it from search results just because it was about a subject with a similar name to that of Takedown AI’s client.
Temara and honeyybee did not respond to a request for comment.
Takedowns AI CEO Kunal Anand told me that the company has filed 12 million takedowns requests to Google since 2022. He said that Takedowns AI uses facial recognition, keyword searches, and human reviewers to find and take down copyrighted content, and said he was overall confident in the company’s accuracy. Anand told me that sometimes his clients use Google Search’s API to see what results come up when they search for themselves, then ask Takedowns AI to remove everything on that list as is, which is what he thinks might have happened with Temara and honeyybee.
“We don't really review it [the list] because we are an agent for them,” Ananad said. “For the requests that we send out ourselves, usually they get reviewed, but sometimes they [clients] do a search by themselves, and they come across some content and they flag it and they're like, ‘We want this taken down.’ We don't review that because that is something that they want taken down. I'm not particularly sure about this case, but that is what happens. What we planned on doing was also reviewing these but it's usually not very fruitful, because the user is very sure they want that claim. And even if we say, ‘Hey, we don't think you should do that,’ they're like, ‘We want to do it. Just do it because I'm paying you for this.’ And if we just say, do it yourself, that kind of takes away the business from us. So that is basically how it works.”
Yvette van Bekkum, the CEO of Cam Model Protection, a company that’s offering the same services as Takedowns AI but that has been in business since 2014, told me that her company does not process requests like Anand described for clients. Cam Model Protection also uses AI, reverse image searches, and keyword searches to find infringing material, but it has systems in place to prevent false positives, Bekkum told me. These include a database of “whitelisted” content that it shouldn’t file takedown requests against, and human verification that each link the company sends to Google actually points to infringing content.
“Just a news article is not a copyright infringement,” Bekkum told me. “If there is no content being used or only a name being named, I don't need to explain to you that it's not in violation. Everybody can make a mistake, of course, but if you just randomly gather [links] and then report it, if it's not grounded on an infringement, you should not report it, of course.”
Bekkum and Ananad both said they understand why creators don’t want to click on every link that might be infringing on their copyright. It’s not only too much work—that’s why companies like Cam Model Protection exist in the first place—it also requires sifting through a sea of pornography they don’t want to see.
“This process is so time consuming,” Bekkum said. “And they do not want to focus on all that negative energy in Googling their name and seeing pages and pages full of links leading to illegal content.”
Elaina St. James, an adult content creator, told me she used a copyright takedown service and that it was most helpful when she flagged offending sites herself. St. James said she used the service to take down pirated content as well as catfishing accounts using her images, a problem 404 Media previously talked to her about. Overall, St. James said these services are useful but imperfect.
“I think they [DMCA takedown request companies] should stop overpromising,” she told me in an email. “There are some platforms—TikTok in particular—that do not comply. Tube sites in foreign countries also rarely comply.”
Automation of DMCA takedown requests has existed for years and has always resulted in some errors. Similar problems have also plagued YouTube’s automated Content ID system for years. More sites are likely to get caught in the crossfire as more content creators strike out on their own and turn to these services in an attempt to protect their income.
It’s an issue at the intersection of several critical problems with the modern internet: Google’s search monopoly, rampant porn piracy, a DMCA takedown process vulnerable to errors and abuse, and now the automation of all of the above in order to operate at scale. No one I talked to for this story thought there was an easy solution to this problem.
“It's all science fiction, but in the dumbest possible way,” Meredith Rose, a senior policy counsel with Public Knowledge who focuses on copyright, DMCA, and intellectual property reform, told me. “At the end of the day, the DMCA takedown provisions are a way to get speech off the internet. That's a very powerful tool. Even if you're not outright malicious, if somebody says something nasty about you and you want to keep your name out of their mouth, the DMCA kind of lets you do it without anybody checking your work. And so it is this really interesting case study in when you build these tools that give the power to anybody, even people who might not be who they say they are in these applications to get stuff taken down. Abuse happens. Sometimes it happens at scale. It happens for all different kinds of reasons. Sometimes it's just malice, sometimes it's incompetence, sometimes it's buggy automation [...] I feel like with AI, we're going to see a lot more of this.”
Anand said he believes the responsibility is with creators.
“The best way to solve that is to educate the creators more that this is not their content,” he told me. “A lot of creators are very scared, and what they want is everything about them taken down from everywhere. And then they start getting more aggressive with their takedowns.”
A spokesperson for Google told me that the vast majority of DMCA removals come from reporters who have a track record of valid takedowns, and that its DMCA removals process aims to find a balance between making it easy and efficient for rightsholders to report infringing content while also protecting free expression on the web.
“We actively fight fraudulent takedown attempts by using a combination of automated and human review to detect signals of abuse,” the Google spokesperson said. “We provide extensive transparency about these removals to hold requesters accountable, and sites can file counter notices if they believe a removal was made in error.”
I feel like the problem mentioned is like a drop in the ocean compared to the enshitification of Google as a whole. Google has been almost unusable even if this didn’t happen.
How do you think DDG 'skips the SEO stuff'? They don't have their own index, they are a meta search engine and the indices they rely on are subject to SEO efforts of the various pages
Google's control over the internet will harm all users as they scram further to find places where profit can be squeezed more and more until they're functionally bankrupt.
not just internet, they are going to prevent android users from installing any programs they dont approve first. So most likely goodbye f-droid and any opensource program
Yeah but really does that even matter when the top results are just ads anyway? The problem is advertising has taken over search engines and now AI makes it even less likely people "searching" for things will even bother to click off of the search website.
DMCA takedown abuse isn't anything new, this article seems like it was just due to 404 media having to deal with it, onlyfans is tangentially related and clearly just used in the headline for clickbait purposes... I really expected better of 404 media, The issue is a valid and increasingly worse one, it shouldnt need a clickbait headline. "DMCA Automation is ruining the internet" or something to that effect would have been a lot better.
This whole thing is also a scam on content creators, people arent pirating content by searching for it on google, they're finding out about websites by talking to people on discord (which itself is not searchable of course) and other such services. Anyone paying for these kind of takedown services is getting taken for a ride.
DMCA takedown abuse isn’t anything new, this article seems like it was just due to 404 media having to deal with it, onlyfans is tangentially related and clearly just used in the headline for clickbait purposes… I really expected better of 404 media, The issue is a valid and increasingly worse one, it shouldnt need a clickbait headline. “DMCA Automation is ruining the internet” or something to that effect would have been a lot better.
That's true, but if the important thing is to draw attention to this issue, this is a good way of doing it even if it's a creative interpretation of the truth.
Enhanced capabilities and availability of tools make a difference. If a company gets the "brilliant" idea to "use AI" to file DMCA requests, encouraged by recent hype of AI and proliferation of many AI tools to make it easy to deploy, then it suddenly becomes a problem, hence why this article exists right now.
But these tools were available for years at least. How do you think Youtube strikes down videos automatically by it's own since like 2012.
The issue is that again and again simpletons are yelling at the hammer instead of the guy that's smashing everything with it. This is so tiring, it makes all of us look so fucking stupid.
It's possible to complain about both because the improved hammer makes the guy smashing everything with it an even bigger problem. Either way it brings attention to the issue, but I have no confidence in politicians making a change for the better at this point, they'll probably just make it even worse.
“We don't really review it [the list] because we are an agent for them,” Ananad said. “For the requests that we send out ourselves, usually they get reviewed, but sometimes they [clients] do a search by themselves, and they come across some content and they flag it and they're like, ‘We want this taken down.’ We don't review that because that is something that they want taken down. I'm not particularly sure about this case, but that is what happens. What we planned on doing was also reviewing these but it's usually not very fruitful, because the user is very sure they want that claim. And even if we say, ‘Hey, we don't think you should do that,’ they're like, ‘We want to do it. Just do it because I'm paying you for this.’ And if we just say, do it yourself, that kind of takes away the business from us. So that is basically how it works.”
They're a cog in a broken machine that needs grease like every other one. It's all just business speak for "why would we want to say something that would make the customer not pay us?"
The DMCA is what you get when you get greedy media and Internet companies to write legislation for lawmakers that have no idea what any of that means, while in the background the people have not had a chance to use any of it. It should have come with an expiration date 10 years from signing and should have been redone now that we are actually using the stuff.
That is the DMCA working exactly as intended. Giving companies the ability to leverage their own wealth and resources to abuse and suppress competition was always the DMCA’s goal.
Copyright in general is about suppressing and abusing competition, there's a little bit of difference now that the old Victorian-style copyright laws lasted as long as the author, more or less, and every legal action was taken through a court, not like these letters of happiness.
It's funny how we seem similar to the pre-WWI mood of "everything has been invented, abolish patents", I wonder if the "pre-WWI" part is too going to rhyme. Hope that not, of course, but most of the innovation seems to be in direct or indirect warfare (all of big tech is honestly that). And there's one nation whose elites seem to make weird destructive moves. And which is on the down trajectory in its GDP relative to the world for the last 50 years. And which has the world's biggest military spending.
After all, humans need a reminder that for the plethora of technologies that seem like a favorable to them weapon unseen before, there are also similarly many technologies that may be unfavorable to them weapons unseen before.
Nazi Germany used radio and encryption and maneuverability and wonderful air force to achieve successes, then the other sides used radars and computers and mass modular production and MLRS'es.
Perhaps the current rotting of copyright and patent system is because the elites think they don't need more natural peaceful development. Global bloodletting usually heals that kind of ideas. Some things can only be learned on your own experience.
Copyright is a surviving instance of the old system of royal warrants: monopolies granted by a monarch, usually to cronies, occasionally as a reward for some kind of good work (scientific discovery, work of art, etc).
It's a system that's full of opportunities for corruption and bureaucratic oppression, and should either be massively scaled back, or dumped entirely. It does far more harm than good.
I agree, unfortunately things only keep existing when there's balance between their sides in power.
Such balance is - those benefiting from copyright have a lot to offer and threaten to those making copyright, and the other way around.
It's all military logic now. 50 years ago it could have been solved by a popular movement, now any movement really threatening copyright will have its figures murdered left and right.
Yup. People talking about DMCA reform, or adding penalties for false takedown requests… In reality, this is the DMCA working exactly as intended. It’s like discussing police reform, but the police are functioning exactly how the ruling class want them to.
Just allow companies to charge a small fee to process a DMCA takedown, and establish a daily compensation rate based on view counts for the uploader (cost payable by the company issuing the takedown - not the entity they represent). Suddenly you only issue a takedown for clear infringement, with the cost paid by the uploader only when clear proof is given that it is a DMCA infringement. If there is a long delay, the uploader gets more compensation, whereas the uploader is only liable for the initial takedown fee.
Can't speak to the why but in practice, nobody ever gets punished for abusing DMCA takedowns even with they use bots to spam fraudulent claims, and it's a widespread practice.
Just because something is legal, doesn't mean it's not scummy.
And as for people getting shit for free, I support a maximalist position on right of first sale: sharing what you own should be legal in all cases. If that inconveniences some mass aggregator of content, tough shit: the ease of sharing gives the lie to the notion that the aggregator adds any value, instead, they're just rent-seeking parasites.
How OnlyFans Piracy Is Ruining the Internet for Everyone
The internet is becoming harder to use because of unintended consequences in the battle between adult content creators who are trying to protect their livelihoods and the people who pirate their content.
Porn piracy, like all forms of content piracy, has existed for as long as the internet. But as more individual creators who make their living on services like OnlyFans, many of them have hired companies to send Digital Millennium Copyright Act takedown notices against companies that steal their content. As some of those services turn to automation in order to handle the workload, completely unrelated content is getting flagged as violating their copyrights and is being deindexed from Google search. The process exposes bigger problems with how copyright violations are handled on the internet, with automated systems filing takedown requests that are reviewed by other automated systems, leading to unintended consequences.
These errors show another way in which automation without human review is making the internet as we know it increasingly unusable. They also highlight the untenable piracy problem for adult content creators, who have little recourse to stop their paid content from being redistributed all over the internet.
I first noticed how bad some of these DMCA takedown requests are because one of them targeted 404 Media. I was searching Google for an article Sam wrote about Instagram’s AI therapists. I Googled “AI therapists 404 Media,” and was surprised it didn’t pop up because I knew we had covered the subject. Then I saw a note from Google at the bottom of the page noting Google had removed some search results “In response to multiple complaints we received under the US Digital Millennium Copyright Act.”
The notice linked to the Lumen Database, which keeps a record of DMCA complaints, who filed them, and for what. According to the Lumen Database, the complaint was filed by a company called Takedowns AI on behalf of content creator Marie Temara. Takedowns AI is one of many companies that help content creators, especially adult content creators, to scan the internet for images and videos they posted behind paywalls on platforms like OnlyFans and posted elsewhere for free. These companies also file DMCA takedown requests and navigate the copyright systems of big platforms like YouTube, Instagram, and Reddit. One of the most effective ways of preventing people from finding this pirated content is sending DMCA takedown requests to Google asking the search engine to delist results to sites that share it. As its name implies, Takedowns AI heavily relies on automation to do this work.
The complaint that impacted 404 Media included a list of 68 links to different websites that allegedly violated Temara’s copyright on content she posted to Instagram, OnlyFans, and other platforms. This was the allegedly offending link on 404 Media, which is a collage Sam made for her AI Facebook therapists story.
The collage includes Meta's AI-generated profile pictures of three of these AI therapists. The story itself has nothing to do with Temara, and the profile pictures look nothing like her. In fact, it would be hard for anyone to claim copyright for that image because in 2023 a U.S. court ruled that AI-generated art can’t be copyrighted.
I went through every other link in the same complaint and couldn’t find even one link that looked like it violated Temara’s copyrights. There were images from Grand Theft Auto V, famous baseball players, robots, stock images of people at theme parks, and movie posters, none of which looked remotely similar to Temara. In addition to 404 Media’s article about AI therapists, some of the pages that Google removed from search results due to this complaint included tech site wccftech.com, horror movie site bloody-disgusting.com, and rugby and wrestling sites.
These links are also just part of one complaint out of hundreds that Takedowns AI files every day. I looked through dozens of complaints to Google filed by Takedowns AI that were archived by the Lumen Database. The vast majority of them appear to be legitimate, but I did find other egregious mistakes. One of the worst mistakes I saw was a takedown request filed on behalf of a creator who goes by “honeyybee” against an article about actual honey bees on the University of Missouri’s website. The takedown request clearly targeted the article and caused Google to remove it from search results just because it was about a subject with a similar name to that of Takedown AI’s client.
Temara and honeyybee did not respond to a request for comment.
Takedowns AI CEO Kunal Anand told me that the company has filed 12 million takedowns requests to Google since 2022. He said that Takedowns AI uses facial recognition, keyword searches, and human reviewers to find and take down copyrighted content, and said he was overall confident in the company’s accuracy. Anand told me that sometimes his clients use Google Search’s API to see what results come up when they search for themselves, then ask Takedowns AI to remove everything on that list as is, which is what he thinks might have happened with Temara and honeyybee.
“We don't really review it [the list] because we are an agent for them,” Ananad said. “For the requests that we send out ourselves, usually they get reviewed, but sometimes they [clients] do a search by themselves, and they come across some content and they flag it and they're like, ‘We want this taken down.’ We don't review that because that is something that they want taken down. I'm not particularly sure about this case, but that is what happens. What we planned on doing was also reviewing these but it's usually not very fruitful, because the user is very sure they want that claim. And even if we say, ‘Hey, we don't think you should do that,’ they're like, ‘We want to do it. Just do it because I'm paying you for this.’ And if we just say, do it yourself, that kind of takes away the business from us. So that is basically how it works.”
Yvette van Bekkum, the CEO of Cam Model Protection, a company that’s offering the same services as Takedowns AI but that has been in business since 2014, told me that her company does not process requests like Anand described for clients. Cam Model Protection also uses AI, reverse image searches, and keyword searches to find infringing material, but it has systems in place to prevent false positives, Bekkum told me. These include a database of “whitelisted” content that it shouldn’t file takedown requests against, and human verification that each link the company sends to Google actually points to infringing content.
“Just a news article is not a copyright infringement,” Bekkum told me. “If there is no content being used or only a name being named, I don't need to explain to you that it's not in violation. Everybody can make a mistake, of course, but if you just randomly gather [links] and then report it, if it's not grounded on an infringement, you should not report it, of course.”
Bekkum and Ananad both said they understand why creators don’t want to click on every link that might be infringing on their copyright. It’s not only too much work—that’s why companies like Cam Model Protection exist in the first place—it also requires sifting through a sea of pornography they don’t want to see.
“This process is so time consuming,” Bekkum said. “And they do not want to focus on all that negative energy in Googling their name and seeing pages and pages full of links leading to illegal content.”
Elaina St. James, an adult content creator, told me she used a copyright takedown service and that it was most helpful when she flagged offending sites herself. St. James said she used the service to take down pirated content as well as catfishing accounts using her images, a problem 404 Media previously talked to her about. Overall, St. James said these services are useful but imperfect.
“I think they [DMCA takedown request companies] should stop overpromising,” she told me in an email. “There are some platforms—TikTok in particular—that do not comply. Tube sites in foreign countries also rarely comply.”
Automation of DMCA takedown requests has existed for years and has always resulted in some errors. Similar problems have also plagued YouTube’s automated Content ID system for years. More sites are likely to get caught in the crossfire as more content creators strike out on their own and turn to these services in an attempt to protect their income.
It’s an issue at the intersection of several critical problems with the modern internet: Google’s search monopoly, rampant porn piracy, a DMCA takedown process vulnerable to errors and abuse, and now the automation of all of the above in order to operate at scale. No one I talked to for this story thought there was an easy solution to this problem.
“It's all science fiction, but in the dumbest possible way,” Meredith Rose, a senior policy counsel with Public Knowledge who focuses on copyright, DMCA, and intellectual property reform, told me. “At the end of the day, the DMCA takedown provisions are a way to get speech off the internet. That's a very powerful tool. Even if you're not outright malicious, if somebody says something nasty about you and you want to keep your name out of their mouth, the DMCA kind of lets you do it without anybody checking your work. And so it is this really interesting case study in when you build these tools that give the power to anybody, even people who might not be who they say they are in these applications to get stuff taken down. Abuse happens. Sometimes it happens at scale. It happens for all different kinds of reasons. Sometimes it's just malice, sometimes it's incompetence, sometimes it's buggy automation [...] I feel like with AI, we're going to see a lot more of this.”
Anand said he believes the responsibility is with creators.
“The best way to solve that is to educate the creators more that this is not their content,” he told me. “A lot of creators are very scared, and what they want is everything about them taken down from everywhere. And then they start getting more aggressive with their takedowns.”
A spokesperson for Google told me that the vast majority of DMCA removals come from reporters who have a track record of valid takedowns, and that its DMCA removals process aims to find a balance between making it easy and efficient for rightsholders to report infringing content while also protecting free expression on the web.
“We actively fight fraudulent takedown attempts by using a combination of automated and human review to detect signals of abuse,” the Google spokesperson said. “We provide extensive transparency about these removals to hold requesters accountable, and sites can file counter notices if they believe a removal was made in error.”
Hey, bit of a crazy idea , but what if we accelerated the destruction of the internet, destroyed it from within before those trying to capture it were ready to take over ?
Yes. An AI alternative to LOIC is desperately needed, for every humanist with a computer to run and to attack the enemy with streams and streams of textual and visual garbage.
I mean like 89% of the shit on Reddit is some OF model advertising.
fr
This
Real
Your comment has been banned and you have been banned from this subreddit for < insert random inapplicable rule >. This message is automated by moderator bot. Do not reply.
Copyright, with a reasonable term (say, 14 years) is potentially useful in preserving author's rights to profit from their own works. Patent laws should be applied far more strictly with respect to prior art and what constitutes a significant innovation, and should never be applied to algorithms. "Same as X, but ON THE INTERNET!" patents should be rejected in every case.
It's monopolistic corporations and incompetent governments that are runing the Internet for everyone. OF is a pox, but compared to those larger factors, OF piracy is nearly irrelevant. It might even be a positive if it damages OF.
In a new paper, several Stanford economists studied payroll data from the private company ADP, which covers millions of workers, through mid-2025. They found that young workers aged 22–25 in “highly AI-exposed” jobs, such as software developers and customer service agents, experienced a 13 percent decline in employment since the advent of ChatGPT. Notably, the economists found that older workers and less-exposed jobs, such as home health aides, saw steady or rising employment. “There’s a clear, evident change when you specifically look at young workers who are highly exposed to AI,” Stanford economist Erik Brynjolfsson, who wrote the paper with Bharat Chandar and Ruyu Chen, told the Wall Street Journal.
In five months, the question of “Is AI reducing work for young Americans?” has its fourth answer: from possibly, to definitely, to almost certainly no, to plausibly yes. You might find this back-and-forth annoying. I think it’s fantastic. This is a model for what I want from public commentary on social and economic trends: Smart, quantitatively rich, and good-faith debate of issues of seismic consequence to American society.
Meh. Nothing in this article is strong evidence of anything. They're only looking at a tiny sample of data and wildly speculating about which entry-level jobs are being supplanted by AI.
As a software engineer who uses AI, I fail to see how AI can replace any given entry-level software engineering position. There's no way! Any company that does that is just asking for trouble.
What's more likely, is that AI is making senior software engineers more productive so they don't need to hire more developers to assist them with more trivial/time consuming tasks.
This is a very temporary thing, though. As anyone in software can tell you: Software only gets more complex over time. Eventually these companies will have to start hiring new people again. This process usually takes about six months to a year.
If AI is causing a drop in entry-level hiring, my speculation (which isn't as wild as in the article since I'm actually there on the ground using this stuff) is that it's just a temporary blip while companies work out how to take advantage the slightly-enhanced productivity.
It's inevitable: They'll start new projects to build new stuff because now—suddenly—they have the budget. Then they'll hire people to make up the difference.
This is how companies have worked since the invention of bullshit jobs. The need for bullshit grows with productivity.
Top upvoted comment just tore a big fat hole into the entire argument and I have to say, good for the comments section. That's so rare.
One open question here is whether we’re seeing youth employment decrease because AI is effectively replacing entry level workers in these fields, or because executives wrongly think AI can or will soon be able to do so?
You have to assume that if anybody puts a hiring freeze for junior employees right now it'd be out of some combination of caution, hype and insecurity about the economic landscape thanks to the usual suspects.
Turns out if the discussion is "quantitatively rich" but is ignoring the obvious qualitative observation it may end up flip-flopping a bunch. I'm not sure I'm as excited about that as the author, because man, is that a constant of the modern corporate world and does it suck and cost people money and stress.
I fail to see how AI can replace any given entry-level SE
You don't "get it"?
AI is making senior software engineers more productive so they don’t need to hire more devs
Yes, I think you do.
I largely look at this as leadership using AI hype as an excuse to cut staff regardless of actual productivity. The house of cards hasn't come down quite yet.
There’s a growing wisdom gap coming in America. The people who are already well versed in company practices and culture are going to use AI to complete the tasks that they would have otherwise given to assistants and junior resources.
The junior resources are going to struggle to find jobs because they are lacking in the KSAs that schools simply cannot provide training for. And that means when us Gen Xers and later Millenials retire there could be a major gap where we have few people with that inherent knowledge to replace us. And where there’s no work and no hope, you get something akin to what is starting to occur in China right now…or revolt.
My hope is that schools will be rethought and there will be a lot more focus on getting an internship early and for the long term. Something more like apprenticeships, which the blue collar workforce maintained, but it’s something we’ll likely need to bring back to white collar jobs.
This isn’t to say that schools should diminish a well rounded education. I think it’s extremely important for students to take electives outside of their focus for a multitude of reasons, one being that it helps students realize the importance of how others contribute to society.
Apprenticeships can help to fill the knowledge gap, but the white collars that are in the jobs now will also need to be retrained and made comfortable to work with a large influx of apprentices to make this approach a success.
There’s a growing wisdom gap coming in America. The people who are already well versed in company practices and culture are going to use AI to complete the tasks that they would have otherwise given to assistants and junior resources.
Counterpoint: no, they are not. Not with the current path of tech progress on the field, at least.
Because seniors well versed in company practices and culture will get tired of having to manually redo junior work corrections really quick, and we are nowhere close to closing the error correction needs at this point.
Repetitive work that could feasibly have been automated or removed already? Maybe. There was a TON of room for automation that people weren't investing on doing and the AI gold rush will feasibly take advantage of some of that. But AI replacing junior jobs wholesale? Nah. The tech isn't there.
Exactly. The senior is willing to put up with the constant questions and mistakes of a junior/intern, because after a few months, they will be better and take some workload off the senior’s shoulders.
With “AI”, there is no learning curve, it’s like you get a different fresh intern every day, and you have to correct the same mistakes constantly.
You really think education reform is going to win out over creating a giant pool of uneducated serfs for Jeff‘s warehouses? I would give anything for your sense of optimism, honestly.
Here we are celebrating Labor Day, the day that celebrates workers rights - overtime pay for working over 40 hours, limiting children from having to work in factories, weekends and time off.
It was a hard fight from serfdom to poor factory conditions to now. We stand on the shoulders of giants.
The "researchers" in no way demonstrate that "AI" replaced any of the jobs lost. They don't even raise the question of whether "AI" is used to discipline labor despite not being much of a productivity aid.
You're right. I could believe these data might be explained by a lot of businesses being in a "wait and see" phase, hiring conservatively while they see how the AI thing shakes out
It's doesnt have to work, it just has to be convincing enough to get the bean counters and/or incompetent/sociopathic upper management to buy in to the idea that they can save money.
Same as always, if the shitstorm created by a decision isn't immediately devastating or can be incontrovertibly tied to said decision then that's just BAU.
but the time the shitshow starts playing the preroll trailers the golden parachutes and bonuses have been claimed.
For them, this isn't broken, this is how the game works.
It doesn't matter if the implementation is good. All that matters is that middle management gets more stuff "done" with fewer people. Where the definition of "done" is just tgat it hasn't exploded in their face.... yet....
The company I work for has replaced a lot of its employees with AI. It's absolutely useless and we have to cover the loss but the fact it doesn't work very well doesn't help the fired employees.
Junior devs and sysadmins who do not much very useful stuff yet, but get some basic experience. And people whose main required traits are human voice and following script.
Transient processes are a thing, one can have plenty of middle and senior devs and sysadmins, with the economy not producing new ones anymore. So the employers are hiring those, and replacing juniors with AI. Whether that works I'm not sure.
So at some point the AI bubble will be over (at least in dev and sysadmin and such work), but there will be fewer developers, and there might eventually be a situation where there are fewer qualified developers in the economy overall. Which would give centralized corporate things a market advantage over smaller non-corporate things, due to cost of development growing after the fall happening now.
While for some not very qualified jobs humans won't be needed anymore - while that "AI" is expensive, it might really be, even after the bubble crash, more affordable than hiring a human (in a western country) for a bullshit job - except in everything I've read those bullshit jobs were treated as social responsibility to teach work ethic to growing generations, that weird mix of individualist and working class themes in books describing pre-Depression USA. Yes, individualism is important and being self-reliant is important, but even that protestant ethic wasn't about capitalism more than it was about dignity and hard work.
I think Silicon Valley is consciously playing Asimov's Foundation with our planet (seeding technologies affecting humanity's development by some schedule with expected global results), except where Asimov's Foundation was about preserving knowledge and civilization, they are moving in the opposite direction. That is, they may not understand it. They may think they are building that sci-fi empire the Foundation begins with. But in actuality they are breaking concrete and steel things that work and replace them with paper huts kinda resembling something that would work better. Metaphorically.
They don't understand what an empire is, neither the "mandate of heaven" kind nor the "unity of civilization" kind (heck, even the Soviet covertly Christian "building the city of sun" kind, like in Vysotsky's song - "... но сады сторожат и стреляют без промаха в лоб"). You don't build an empire by burning libraries and poisoning discourses, you also don't build an empire by making every its citizen uncertain whether they are a free man or a slave (it's a common misconception to start an attempt at an empire from points where previous empires failed ; that state is usually expected to fail again for the same reasons).
as a consultant/freelancer dev whose entire workload for the past year has been cleaning up AI slop, no with dev it hasn't been what I would say a smooth or even good implementation. for my wallet? been a fantastic implementation, for everyone else? not so much.
The thing is as a TOOL it's great depending on the model. As a rubber duck? fantastic. As something that the majority of companies have utilized with vibe coding to build something end to end? no, it's horrible. It can't scale anything, implements exploits left right and center, and unlike junior devs doesn't learn anything. If you don't hold its hand during a build then it'll quickly go off the rails. It'll implement old APIs or libraries or whatever simply because those things have the most documentation attached to it.
An example. a few weeks ago a client wanted to set up a private git instance with Forgejo. They had Claude Code set it up for them. the problem? Claude went with Forgejo 1.20. ForgeJo is currently on 12.0. MASSIVE security hole right there. Why did Claude do that? 1.20 had more documentation as opposed to 12.0. And when I say "documentation" I could simply be referring to blog posts, articles, whatever that talked about it more than the latest version because The LLM's will leverage that stuff when making decisions for builds. You also see it if you want something in Rust+Smithy. Majority of the time the AI will go for a very outdated version of Smithy because that's what a lot of people talked about at one point. So you're generating massive tech debt before even throwing something into production.
Now like I said as a tool? a problem solver for a function you can't figure out? it's great. the issue is like I said companies aren't seeing it as a tool, they're seeing it as a cost saving replacement for a living human being which it is not. It's like replacing construction worker with a hammer attached to a drone and then wondering why your house frame keeps falling over.
I've seen some beginners trying to use AI to generate simple features for visual novels using Sugarcube (Twine). The code was awful, they didn't understand it and even people with some experience needed time to understand what the fuck the code was even trying to do.
It doesn't have to be a good implementation, it only has to be good enough for a demo to get the C-suite saying "Oh, slap a chatbot on there and then fire half the department that handles this now".
Putting AI to work in place of actual humans in help desk/customer service sounds like a natural evolution, since instead of being ignored by a dumb, exhausted human who's perpetually wishing to find a better job and has a 80% chance of being useless ("I will transfer to the appropriate person"), you'll be misunderstood by a robot.
Just wait all this is just a scam without any real revenue. That's why it's just investments right now into big companies. They're already starting to fail not really recognizing AI isn't supposed to be used in that way anyways.
Scientists say the planned decommissioning of a valuable research vessel is part of a series of actions by the Trump administration that take aim at climate science.
Zimbabwe published draft regulations to establish a National Climate Fund that will finance projects aimed at mitigating the impact of climate change and respond to emergencies.
One is free floating and the other is attached to the international space station. The original team hopes to find funding to save the ISS attached monitor. A more detailed article below
A group that infiltrated Microsoft’s headquarters building this week disputed the company’s account of the incident — describing their sit-in as nonviolent and saying the “listening devices” allegedly left behind were phones that fell from their pockets when they were arrested.
“As Brad himself admits, if someone were to plant listening devices, this is not how they would do it,” said Hossam Nasr, one of the leaders of the group No Azure for Apartheid, referring to comments made by Microsoft President Brad Smith after seven members of the group occupied his office Tuesday afternoon. “If anything, we would like our phones back, please.”
The group, which is calling on Microsoft to cut ties with Israel over the alleged use of its technology against Palestinians in Gaza, also disputed the company’s assertion that its members do not represent elements of its workforce, and questioned the sincerity of Microsoft executives in addressing the issues the protesters have raised.
Un’altra Mostra del Cinema si accende con una presenza che fa discutere e riflettere: quella di Kevin Spacey. L’attore è tornato a calcare un red carpet internazionale di grande rilievo, scegliendo il Lido di Venezia per segnare un ulteriore passo nel suo atteso ritorno sotto i riflettori, dopo la conclusione positiva delle sue vicende legali.
La famiglia Parr sta per tornare, ma armati di pazienza! I fan de Gli Incredibili dovranno aspettare ancora un po’ per il terzo capitolo. L’annuncio ufficiale è arrivato, ma la data di uscita è ancora lontana: 2028, se tutto va bene.
Snow Lemmy
in reply to ro. • • •