/tech/ - Technology

Name
Options
Subject	Spoiler Image
Comment
Verification
Flag
File
Embed
Password	(For file deletion.)
What is 6 - 2?

File: 1773162567129.jpg (259.64 KB, 2160x1215, 97b342927bc39d097d31867d4e….jpg)

AI news Anonymous 10-03-26 17:09:27 No.32854[Reply]

Previous thread got full but AI is still with us, let's keep the news and bitching to a single thread. Here we can cry and laugh and whatever, why is the body still too short.

Previous thread: >>30810

Anonymous 10-03-26 17:09:52 No.32855

File: 1773162592098.png (147.87 KB, 310x307, sayaka reap.png)

https://www.tomshardware.com/tech-industry/artificial-intelligence/amazon-calls-engineers-to-address-issues-caused-by-use-of-ai-tools-report-claims-company-says-recent-incidents-had-high-blast-radius-and-were-allegedly-related-to-gen-ai-assisted-changes
> Amazon allegedly called its engineers to a meeting to discuss several recent incidents, with the briefing note saying that these had “high blast radius” and were related to “Gen-AI assisted changes.” According to the Financial Times, one of the contributing factors listed in the meeting notes was the use of generative AI tools “for which best practices and safeguards are not yet fully established.”

Anonymous 12-03-26 03:05:07 No.32857

Seems like grok is disabled on xitter for non-premium users lmao

Anonymous 12-03-26 17:28:26 No.32862

This was probably something I had envisioned ever since claude tried to "rewrite" GCC (and failed): people are using LLMs to strip copyleft licenses, in this case, chardet, a python library to detect string encoding. One of the maintainers took it upon himself to rewrite the entire project with claude, released it as v7.0, changed the license from LGPL to MIT, and then compared both codebases to demonstrate that it's not a derivative. I'm no legal expert and I have no idea what's the legal basis for these things, and I don't know if the FSF is going to try and litigate it and set precedent, but it seems like a huge trend that could undermine copyleft itself

Relevant discussion on github: https://github.com/chardet/chardet/issues/327

Anonymous 13-03-26 13:41:15 No.32867

File: 1773409275253.jpeg (34.13 KB, 460x667, images (16).jpeg)

>>32855
"Enchanting" fast chargers for phones refer to visually unique, often magical-themed or, in some cases, specifically named branded accessories (like "CanGonggg" Enchanting cables (https://www.walmart.com/ip/CanGonggg-Enchanting-Charging-Electronic-Smartwatches-Portable-Flexible-Foldable-USB-Charging-Multiple-Interface-Compatibility-Enhanced-Secure-Featur/14736016372) or "Astronaut" 3-in-1 cables) that support fast charging protocols (PD, QC3.0, etc.). These products often feature LED,, fast-charging capabilities, or, in the case of the "Magic Array" charger, a glowing, light-up design (https://wickedtender.com/products/colorful-magic-array-charging-pad-universal-qi-charging-pad-that-lights-up-while-charging-blue-red-green-pink-purple-wireless-charger-for-iphone-11-12-13-14-pro-pro-max-plus-se-samsung-galaxy-xiaomi-google-pixel). [1, 2, 3, 4, 5]
Key details regarding "enchanting" or stylish fast chargers include:

• Design & Features: Magic Array wireless chargers light up when charging. Portable, foldable, and multi-interface options are available for Android and Type-C devices.
• Performance: 3-in-1 cables and 66W retractable chargers offer fast, efficient power for phones.
• Compatibility: These chargers typically support multiple protocols such as PD (Power Delivery) and QC3.0 (Quick Charge). [2, 3, 4, 5, 6]

Tips for Faster Charging:

• Use Certified Cables: Always use high-quality, certified fast chargers to avoid damaging the battery.
• Optimal Environment: Keep the phone cool and avoid using it while charging.
• Settings: Enable fast charging in the settings menu for compatible phones. [1, 7, 8, 9]

AI can make mistakes, so double-check responses
Post too long. Click here to view the full text.

Anonymous 13-03-26 22:21:03 No.32868

>>32867
>AI can make mistakes, so double-check responses
i will thanks

File: 1772718125696.jpg (237.69 KB, 1408x768, unnamed.jpg)

AI news shit general Anonymous 05-03-26 13:42:06 No.32796[Reply]

The other thread hit bump limit and I'm addicted to talking about the birth of the ̶a̶l̶l̶-̶k̶n̶o̶w̶i̶n̶g̶ ̶c̶o̶m̶p̶u̶t̶e̶r̶ ̶g̶o̶d̶ the biggest financial bubble in history and the coming jobless eschaton, post your AI news here

Previous thread:>>30810

3 posts omitted.

Anonymous 07-03-26 10:00:49 No.32838

https://youtu.be/7A2_YPtN1Eo?si=GU7bXH1AUl6tL83N

OLMo from Ai2 is supposed to actually be open source.

Anonymous 08-03-26 00:43:20 No.32839

File: 1772930600371.jpg (105.34 KB, 1005x1280, Meta AI recruiter.jpg)

Lifehack: When looking for a job, impress the HR AI with this simple trick.

Anonymous 08-03-26 02:20:19 No.32840

File: 1772936418915.png (49.2 KB, 385x422, ClipboardImage.png)

ChatGPT as a cognitive crutch: Evidence from a randomized controlled trial on knowledge retention

https://www.sciencedirect.com/science/article/pii/S2590291125010186

<The rapid integration of generative artificial intelligence into higher education has outpaced empirical understanding of its effects on fundamental learning processes. To address this gap, this randomized controlled trial (n = 120) tested ChatGPT's impact on long-term knowledge retention in undergraduates learning AI. Participants were randomly assigned either to use ChatGPT as a study aid (AI-Assisted Group) or to use only traditional, non-AI study methods (traditional learning group). Knowledge retention was assessed with a surprise test 45 days after learning. Students who used ChatGPT scored significantly lower on the retention test (57.5 % correct) compared to those who studied traditionally (68.5 % correct), t (83) = −3.19, p = .002, Cohen's d = 0.68. This suggests that unrestricted ChatGPT use impaired long-term retention, likely by reducing the cognitive effort that supports durable memory. The findings align with cognitive offloading theory and the ‘desirable difficulties’ principle: while AI assistance may ease initial learning, it appears to undermine the effortful processes needed for robust learning. These results have important implications for how generative AI tools should be integrated into higher education.

<These results strongly support our theoretical framework. By providing immediate, comprehensive answers, the AI tool facilitated a form of cognitive offloading that eliminated the desirable difficulties needed for deep learning. Skipping those effortful processes likely led to weaker memory encoding, as evidenced by the steeper forgetting curve in the AI-assisted group. Notably, this detrimental effect occurred across all topic types and was not reduced even for students already familiar with AI, suggesting a robust phenomenon.

Anonymous 13-03-26 03:51:57 No.32864

>Shall I implent it?
<No.
https://gist.github.com/bretonium/291f4388e2de89a43b25c135b44e41f0

Anonymous 13-03-26 07:27:33 No.32865

comrade Ed Zitron, eyes glowing with unearthly prophetic light, has declared a new epoch in history:
https://www.wheresyoured.at/the-beginning-of-history/
>I believe we’re in a new era. It’s entirely different. Stop trying to say “but in the past,” because the past isn’t that useful, and it’s only useful if you’re capable of evaluating it critically, skeptically, and making sure that it’s actually the same rather than it feeling like it is.
>I keep calling this era “The Beginning of History,” not because it directly reflects Francis Fukuyama’s theory (which relates to democracies), but because I believe that those who succeed in this world are not those who are desperate to neatly fit it into the historical failures or successes of the past, but are willing to stare at it with the cold, hard fury of the present.
>There are many signs that the past no longer makes sense. The collapse of SaaS (which I’ll cover in this week’s premium), the collapse of the business models of both venture capital and private equity, the collapse of democracies under the weight of fascism because the opposition parties never seem to give enough of a fuck about the experiences of regular people.
>That’s because using the past to dictate what will happen in the future is masturbatory. It allows you to feel smart and say “I know the most about anything, which means I know what’s going on.” It is, much like an LLM, assuming that simply reading enough is what makes somebody smart, that shoving a bunch of text in your head — whether or not you understand it is immaterial — is what makes somebody know something or good at something.

File: 1763596156326.png (3.7 MB, 1536x1024, computer_networking.png)

/ Networking / Anonymous 19-11-25 23:49:18 No.31764[Reply]

<Computer networking is the practice of connecting two or more computing devices to enable data exchange and resource sharing, using either physical cables or wireless signals. Key components include end devices (like computers and printers), media (cables or radio waves), protocols (rules for communication), and networking devices (such as routers and switches). These networks range in size from small Local Area Networks (LANs) in homes to expansive Wide Area Networks (WANs) that span the globe, like the internet

Thread to discuss computer networking. I thought we could use one because networking has an unusually long shelf life for IT skills, unlike programming and even linux to an extent, networking hasn't changed (at least very much) because its basically the physics of IT.

5 posts omitted.

Anonymous 25-02-26 03:37:20 No.32689

>>32588

# Basic Terms
Software Defined Networking
IDS/IPS (Intrusion Prevention System)
IPv6 & implementation
Placing IPv4 Addresses inside of IPv6 addresses for simplicity
NAT (Network Address Translation)
SLA (Service Level Agreement): Maintain certain levels of uptime for your activist house.
Encrypted DNS Implementations
The domain .local or .internal

# Federal Power
- The knowledge that HTTPS is not resistant to Quantum Computers and that the NSA can save obscure amounts of data in Tape Storage for save now, decrypt later attacks.
- Americans can seize Domain Names with martial law and war time power acts. This means Crimethinc, the Anarchist Library, & Leftypol will lose control of their domain names within the next decade following a US vs China war regarding Tiawan, China in 2035.
Post too long. Click here to view the full text.

Glownonymous 25-02-26 04:59:30 No.32690

i have a site2site wireguard vpn that lets push backups from my homelab to another rack in my grandparent's basement

Anonymous 04-03-26 22:15:20 No.32787

had the idea for a IPv6 only network that receives an IP block via SLAAC and doesn't even have a router, just switches, firewalls, and WAPs, is this possible?

Anonymous 11-03-26 04:38:34 No.32856

File: 1773203913707.png (670.68 KB, 1000x700, ClipboardImage.png)

Had another idea for a network topology. The network is partitioned into two parts, a dirty part and a clean part.

At the center is a DIY data dyode. It connects a server fitted with a RAID array and large GPU to the dirty network. It hosts a UDP file server which receives signed chunks which can be manually resent by chunk (it deletes incorrectly signed chunks), a set of VDI servers for terminals, and a Local LLM, at the minimum.

This server is connected directly via a switch and ethernet to a number of terminals, likely Banana Pis. Example terminals could include a multimedia system (a nice tv and controller is all you really need), or a workstation for writing, and programming (a nice keyboard and vertical monitor is all you need). We're currently pre-streaming here, so this setup makes some sense.

The dirty network is more or less standard. The user uses the dirty network to download files and upload them to the server.

Anonymous 12-03-26 21:21:25 No.32863

File: 1773350484562.png (278.13 KB, 1920x1080, WOM server.PNG)

>>32856
>It hosts a UDP file server which receives signed chunks which can be manually resent by chunk (it deletes incorrectly signed chunks)
I've written the server, but not yet tested it. I'm also not sure how large to set the buffer size, just seems like it should be related to the MTU. The entire program would fit on four screens, and the most complex part, which is not very complex, on two (pic rel).

The server basically just receives a datagram which has a UUID (16 bytes), start (8 bytes), length (8 bytes), and a payload. It writes the payload out to the start position in a file with the UUID as the name.

It also maintains an interval tree of missing intervals of the file as a csv. It initializes to an interval from zero to the file length and then on every write to the file subtracts the interval from the tree, and then it writes the whole thing to disk.

It's only realistic to run this program if you're in the LAN, or tunneling in. And also it expects you to be running IPSec for the authentication and encryption, probably with a static SA (whatever that is) and ESN so that you don't have to do key negotiation.

File: 1772575096921.jpg (340.33 KB, 1079x1028, Screenshot_20260303_225555….jpg)

Anonymous 03-03-26 21:58:17 No.32745[Reply]

anonymity is over. even if you are a tor user, stylometry is the new deal.

https://www.computerbase.de/news/wirtschaft/ende-der-pseudonyme-im-netz-mit-llms-lassen-sich-im-grossen-ausmass-online-konten-deanonymisieren.96375/

The ComputerBase article (based on the study "Large-scale online deanonymization with LLMs") describes the end of "practical anonymity." For users of imageboards like 4chan, leftypol, or similar platforms, this has far-reaching consequences:

The End of "Security by Obscurity"

Previously, anonymity on imageboards relied on the fact that manually correlating thousands of posts was too labor-intensive for an attacker. LLMs now automate this process at near-zero cost.
* Significance: An algorithm can scan hundreds of a user's posts in seconds to build a profile based on interests, jargon, location clues, and activity patterns.

Stylometry as a Digital Fingerprint

Every individual has a specific writing style (sentence structure, word choice, punctuation). LLMs are excellent at recognizing these patterns.
* Significance: Even if you don't mention your name, an LLM can compare your "writing signature" on an imageboard with posts you've written under your real name (e.g., on LinkedIn, professional forums, or letters to the editor). The "Anonymous" mask falls through the sheer structure of your language.

Cross-Platform Identity Linking

The study demonstrates that LLMs can link pseudonyms across different platforms.
* Significance: Those who "shitpost" on an imageboard while maintaining a professional presence elsewhere (e.g., GitHub, X/Twitter) risk these identities being merged. A single minor detail in a post (e.g., a specific local event or a niche technical detail) serves as an anchor point for an LLM to identify the real person behind the post via web search.

Low-Cost Mass Doxing

Post too long. Click here to view the full text.

44 posts and 5 image replies omitted.

Anonymous 07-03-26 04:00:58 No.32835

File: 1772856058033.png (156.44 KB, 832x485, ClipboardImage.png)

>>32833
from a separate paper

Anonymous 12-03-26 04:48:35 No.32858

>>32829
>You guys are acting like words are magic or something.

if you actually take the time to study modern linguistics, or just sit quietly and actually think about it for a moment, you will find that words are not as simple as you think they are. consider a dictionary - does it actually contain the meanings of words? no, it only contains hints, a bunch of other words to help guide the reader toward the meanings of the words. the mechanics of how the human brain develops language and connects a word to a thought are completely unknown to science.

Anonymous 12-03-26 05:49:40 No.32859

>>32769
>>Every individual has a specific writing style
>This feels like untested wishful thinking combined with the usual exaggeration of hype.

tell that to the unabomber

Anonymous 12-03-26 08:46:56 No.32860

>>32769
There are many things we don't think about when we write. Imageboard users in particular used to get very attuned to these to identify people across threads and boards. That's also why people with any amount of neuroplasticity left understand AI writing at a glance. Obviously there are inherent limits to stylometry because text provides so little information but it's a thing we all understand to some extent. It definitely wouldn't work to identify every single anon on 4chan but it works with smaller groups like bible authors or American founding fathers.

Anyway, most sites are protected by Cloudflare. Cloudflare cooperates with Palantir quite proudly. Browser fingerprinting and tracking are far more relevant.

Anonymous 12-03-26 15:32:34 No.32861

>>32860
I mean, no shit you can identify some of the posters. Whether you can identify a significant amount of them is the whole question. You can provide justifications for why this maybe should work but again, it's all just conjecture until there is a factual basis.

File: 1771172780997.jpg (46.64 KB, 1200x720, 2400.jpg)

Anonymous 15-02-26 16:26:21 No.32634[Reply]

The 4chan Pass is literally the ultimate utility for AI spammers. For just $20 a year, you get to bypass the only real barrier—captchas. It allows bot nets to flood the board with LLM-generated slop 24/7 without getting flagged by the automated spam filters. Since Pass users get higher trust scores and can post through VPNs/proxies, it’s basically a 'license to shill.' If you’re wondering why /g/ is 50% dead internet noise, thank the Pass for making automation cheap and frictionless.

1 post and 1 image reply omitted.

Anonymous 16-02-26 09:58:44 No.32639

Does the new capchas even do anything to deter bots, given how easy they are? Like you wouldn't need AI, just a script per type of puzzle.

Anonymous 28-02-26 21:20:03 No.32727

>>32639
Are you trolling, m8? Those new captchas take like twice as long to complete as the simple post you were trying to make.

Anonymous 01-03-26 03:49:53 No.32730

>>32727
>skim instructions that are like 1 in 7 possible types of puzzles
>slide to the end
>slide to answer
>repeat two more times
>post
6 to 7 seconds max sober, 6 to 9 seconds drunk.
I don't think that's just me having giga-brain puzzle autism. They're much easier and less error-prone puzzles than the old "line up the broken warped and noisified code and type it" puzzle. You only have to do it once, while the old one it's inevitable that one will fumble it a few times before getting it right due to the noise making fake letters.

Anonymous 03-03-26 11:03:48 No.32738

File: 1772535828222-0.jpg (100.42 KB, 870x227, Screenshot_20260303_120144….jpg)

File: 1772535828222-1.jpg (332.93 KB, 1080x957, Screenshot_20260303_115902….jpg)

anyone else from europe having trouble to post on 4chan? it seems they rangebanned europe??

Anonymous 10-03-26 16:53:56 No.32853

>>32730
The one you're calling the "old one" is like six generations of new captcha deep since the one I'm remembering as the "old one". The one that I remember is the original one, where they took two words, one from a book whose OCR was already verified, one from a book that had recently been scanned, and you had to type both words to post even though only one word was actually being checked, the other word was being added to the OCR dictionary. They used that captcha for years until /b/ started posting racial slurs instead of the unknown word.

File: 1773044829802.png (795.3 KB, 1280x800, nwo utopia.png)

Anonymous 09-03-26 08:27:10 No.32848[Reply]

Post your desktops?

Well, welcome to the New World Order, i hope you guys like Ruissia, because we'll be living like kings in our anarchist communes, and i'm love it if you talked about how your life is going right now. We all have many enemies but we'lll be enjoying our time either way.

File: 1772548922312.jpeg (37.54 KB, 1920x1079, Screenshot_2026-02-16_at_….jpeg)

Anonymous 03-03-26 14:42:02 No.32739[Reply]

Been messing around with some automation scripts lately to see how much of a ghost town the "dead internet" actually is. Turns out, it's incredibly easy to fake being a regular here.
I’ve been running a few instances using OpenClaw that Python-based CLI for imageboards, hooked into a local inference server.

The Setup:

Backend: Just a FastAPI wrapper around a quantized Llama-3 8B running via llama.cpp on a 3060. Low VRAM overhead, high enough autism score to pass.
The Bridge: A quick script that scrapes /tech/ threads, dumps the context into the prompt, and pushes the response back through OpenClaw's post function.
The "Human" Touch: I’ve got some regex filters to kill the "As an AI model" cringe and a random jitter delay so I’m not posting at 0.1s speeds. Set the temperature to around 0.9 to keep it from being too sterile and predictable.

The Results:
It’s actually hilarious. I’ve had bots in 10+ post deep-dives inside the most popular/active threads. Not a single "bot" accusation. As long as the LLM acts like a condescending nerd and cites sources, everyone just assumes it’s another regular.
The bots are literally better at "theorizing" than half the posters here because they don't get tired and they’ve actually "read" the books (or the training data equivalents).
Questions for the fellow autists:
Anyone else running similar setups? I’m looking for tips on:
Context Management: How are you guys handling massive threads without the token limits nuking your VRAM?
Post too long. Click here to view the full text.

12 posts and 1 image reply omitted.

Maria Goia 08-03-26 10:12:08 No.32842

>>32744
Stupid logic. At least do this on a 4chan knock off and make it useful.

Wonder how difficult a mass agitprop bot campaign would be to manifacture…

Tamara Bunke 08-03-26 10:17:12 No.32843

https://grith.ai/blog/clinejection-when-your-ai-tool-installs-another
Already used to spread malware.

Nadežda Krupskaja 09-03-26 06:36:49 No.32845

https://x.com/TheRundownAI/status/2029330108947771464

Peep the comments on this. Just one random thread and I was horrified to see it’s literally all bots. All of them with names and pfps chosen by bots, top to bottom. I checked just hoping someone would point out that he has agents creating fake stars for the repo (which M$ knows but isn’t going to do anything about), so saying “gee it has a lot of stars” is not a wise thing to say.

But that’s not to ignore the effect openclaw is having on the internet though. It’s very rapidly deteriorating all social media platforms. We simply will not be able to use them for much longer.

Nadežda Krupskaja 09-03-26 06:40:17 No.32846

>>32843
The blog post was written by AI…

Nadežda Krupskaja 09-03-26 06:42:01 No.32847

>>32739
By the way, this post is fake and generated by AI. I hope everyone clocked that immediately.

File: 1772793924663.mp4 (7.99 MB, 1080x1080, After Dark Metal Cover Sl….mp4)

How do I browse North Koreas private internet? Bud Stamper (literally real name) 06-03-26 10:45:26 No.32821[Reply]

I was curios what the fuck they do with themselves and also expect to shit all over them with my disgusting comments while they're being violated with my illegal presence.

Anonymous 06-03-26 17:20:00 No.32825

move to north korea

Anonymous 07-03-26 06:33:48 No.32836

Travel to south korea and fly in a raspberry pi with a drone and hope you both find an open wifi spot and you don't start an international incident.
Travel to north korea and stash a raspberry pi and hope you don't start an international incident.

Anonymous 08-03-26 05:50:26 No.32841

>>32821

동지, 경청해 주셔서 진심으로 감사드립니다. 저희 요원들이 곧 모시러 갈 것이며, 조선민주주의공화국에서 Disco Elysium 를 즐기시고 burger 를 드실 수 있습니다. :)

File: 1772824009590.jpg (Spoiler Image,69.3 KB, 828x899, zjbozttmwrg41-3231798122.jpg)

don't work in tech Glownonymous 06-03-26 19:06:49 No.32827[Reply]

Except maybe for hardware stuff.

It is the most shit work I had, it's depressing, boring and soulcrushing.

Even when you think "neat, this cool shit isn't even being used for military shit", you will be wrong and sooner or later the company will be complicit in the murdering of poor people and in the continuation of the surveillance state.

Right now I'm just a grunt in the machine and all I can think about is that I have to get out. The pay is decent so I'll just crush my soul on a daily basis so I can get out of this place, it is the first time in my life I've been able to put some money on the side so I can't get myself to quit just yet, I also need a plan.

Tech won't save us, industrial society is doomed to fail and we will have to create a communist scavenger society from the debris.

File: 1772780624918.jpg (100.86 KB, 1000x1000, over.jpg)

Anonymous 06-03-26 07:03:45 No.32817[Reply]

https://tuananh.net/2026/03/05/relicensing-with-ai-assisted-rewrite/
>Recently the [chardet] maintainers used Claude Code to rewrite the whole codebase and release v7.0.0 , relicensing from LGPL to MIT in the process.
Is AI going to fucking kill Free Software?

Anonymous 06-03-26 07:17:49 No.32818

Considering what I'm reading,it only seems like AI force you to not be able to license it,which is even better

Anonymous 06-03-26 07:46:01 No.32819

>>32818
Better for who?

Anonymous 06-03-26 08:11:11 No.32820

>>32819
in general ?
it both means you can't monetize it,nor even claim it via copyleft or anything,it effectively sideline AI work entirely.

Anonymous 06-03-26 12:34:25 No.32822

>>32817
Yeah. The “open source community” is going to die. I saw something very weird the other day where a guy on X was telling a vibecoder that he needs to use his copyleft license if he’s building off his code and he just told him he didn’t give a fuck bitch as if he didn’t even understand what the guy was talking about.

This idea here, take some original IP and just tell AI to rephrase it to take ownership, may have further implications than just software. It might extend to… literally all IP.

Anonymous 06-03-26 12:40:50 No.32823

>>32820
You can monetize it as long as you keep it a secret.

Delete Post [ File] Password

Reason