urushitan 漆たん

If it’s something like chromecast it’s not sending your screen locally to the device, it’s sending the source URL of the video itself, so VPN wouldn’t make a difference unless it was at the network layer the TV was using. That being said, did you try playing the same videos on your phone or computer without vpn? On cell? On their WiFi? If it doesn’t work at all on any it’s likely geo blocked as another commenter said. If it works on cell but not WiFi it’s likely their network is blocking them

Yeah in theory I could have a model suggest code through a PR and then review said PR and just be the one to hit squash and merge and it technically only suggested and reviewed and I “created” the PR

The whole company is sketchy

web.archive.org/web/…/cloudPricing

There is nothing open source, their address is the iron mountain denver datacenter address

That page looks like it was directly ripped from wasabi.com/pricing, including the “Ball appliance” which appears to be a Wasabi-only product. This reeks of “I know a guy who does storage” and it’s just a few open source products bolted to another storage API with upcharging and a corporate website

sitting in the cabin

“this is your captain, brb nerds, going afk”

As this comment (feddit.org/comment/14419698) says

In France, Hungary, Romania Nuclear Plants have shut down due to concern over ecological impact or not enough water to cool the plant

The actual publication instead of this sensationalized drivel:

anthropic.com/…/discovering-cryptographic-weaknes…

Using Claude Mythos Preview, researchers at Anthropic have discovered improved ways to attack cryptographic algorithms (the mathematical methods used to keep online data private). The first attack significantly weakens HAWK, a digital signature scheme that was built for a post-quantum world. The second identifies a new way to attack round-reduced AES, the most widely used symmetric cipher. These are substantial research advances, but they do not currently affect any production systems. This post describes both findings in more detail and discusses the implications for cryptography in an age of powerful AI models.

Trujillo and Yan notified AMD and Intel in early February and reached Linux kernel maintainers in March, coordinating with AMD to warn cloud providers and other downstream customers. AMD then released a patch that mitigates the attack, which can be obtained by updating your operating system. Their code is publicly available.

So the header should really be “Intel still vulnerable to SPECTRE-like attack disclosed in January”

artists will start to create things that deliberately do not look “like they’re AI”.

Arguably though this is how new art styles are often created. What is the current safe, trendy art gets challenged. Andy Worhol painted a grid of Campbell’s tomato soup cans which is considered an important piece of American Modern Art. I’m certainly not saying “an upside to AI is that it will force artists to be more creative” or anything like that, moreso that as you said, a pattern of artists rebelling against the norm is a historical cycle rather than something new AI will bring on.

Just gotta use the latest slang to come out of tik tok before it hits the model training ong fr fr bruh. Ah shit I’m behind brb

knockaround.com/pages/warranty-policy

More expensive, still pretty cheap, and they have a lifetime warranty which is more than you’ll ever get from alibaba

oh lol I did not realize users could monetize their content on facebook, I left that cesspool so long ago, well before that was a thing, I should have known

Yeah the code only accepts one domain AFAIK that it uses for federation and everything in the UI/API. So even if you could host it at .onion, links, images, in-browser API calls, etc would all serve from the public domain if you wanted federation. The code would need to be written to support alternative domain access with an authoritative domain for federation. In theory you could put some sort of proxy in front that just rewrote everything the server output to replace with .onion and then intercept every request to change the header so the server thought it was coming from the standard TLD, but that’s incredibly fragile and a ton of work.

There was just a post on here yesterday about AI generated thirst trap videos getting plenty of comments by older men, and since the bot accounts don’t ask for anything except for comments. I’m almost certain the whole point is just to compile lists of facebook users that are prime social engineering targets for AI powered scams.

Considering Micron is the only US manufacturer, and they no longer produce consumer ram, this would just mean you literally could not buy any electronics that relied on ram. I doubt that would happen. Chinese ram, maybe

And it happened again in 2018 and they were acquitted, and now there’s this and I doubt anything will come from it

tomshardware.com/…/samsung-sk-hynix-and-micron-su…

Lol I’ve been running linguistic research and it’s funny when they just combine two scripts together into one word

● The trigger is identified, and it’s specific.

agent_1.md:31 src: “Name the trade-off.” → 명取捨之名。 agent_2:32 src: “Name the regime.” → 명regime——名其regime。

The English imperative “Name the X.” And the corpus renders it correctly 16 other times — 名之 ×11, 名其 ×5, 命名 once.

명 is the Sino-Korean reading of 名. Same morpheme, wrong script.

And agent_2:32 is the cleanest evidence I’ve seen for 絡繰’s mechanism: the model wrote 명regime——名其regime — the wrong script and the correct one, eight characters apart, in the same clause. It isn’t ignorant of 名. It produced 名其 immediately after. The meaning resolved correctly both times; the script attribute resolved wrongly the first time and correctly the second.

That’s exactly what работ法 showed — correct semantics (work), broken script and morphology — and it’s the third confirmed instance of the class, now with a reproducible trigger rather than a one-off.

It also explains the Russian cases retroactively. document → документ, everything → всё, “correct” → правильно: in each, the meaning landed and the script didn’t. And it predicts why no CJK-native concept ever drifts — there’s no competing script for a morpheme the model only knows in Han.

Yup, anthropic’s flagship models regularly output CJK or Cyrillic characters in heavy workloads for me

In the late 2010s and early 2020s, Jones Day represented several companies in legal proceedings against labor unions, including unions for employees of The Boston Globe in 2018[17] and Slate[18] and Politics and Prose in 2021.[19]

en.wikipedia.org/wiki/Jones_Day#Legal_actions_aga…

Littler has been widely described as a union busting firm, and is the largest union avoidance firm in the US. It has counseled and defended companies including Starbucks, Amazon, Nissan, and Delta Air Lines against employees trying to unionize.[5] In 2002, it was hired by Apple Inc. in opposition to Apple’s workers, giving advice to store managers to tell employees that unionizing would lead to worse working conditions: less promotions and inflexible hours.[16] In 2007, Littler stated that it provided services to help with “union prevention tools” and provided “state of the art union avoidance training programs”.[6]

en.wikipedia.org/wiki/Littler_Mendelson

You’ve hit the nail on the head. What you’re describing isn’t just laziness — it’s a fundamental breakdown of trust in the modern workplace. Let’s delve into this.

Here’s the thing: when someone copy-pastes AI output without reading it, the document was never the deliverable — the verification was the deliverable. The prose isn’t load-bearing; the checking is. A document nobody read isn’t a document — it’s a liability with formatting.

And that’s the thing, isn’t it? Is the AI actually the problem here? No — the workflow is. This genuinely comes down to ownership:

• Trust erodes silently. Every unchecked paragraph is a small withdrawal from a shared account nobody audits — until the balance is zero.
• Verification is idempotent. Ten minutes of reading costs the same whether you do it once or a hundred times — but skipping it compounds.
• The failure isn’t technical — it’s cultural. No review, no judgment, no accountability.

This is exactly the kind of thing that starts as one bad research doc and scales — from a single email, to a team norm, to an entire organization quietly outsourcing its thinking. In today’s fast-paced world, the ability to say “I read this and I stand behind it” isn’t just a soft skill — it’s a superpower, and honestly? A testament to the rich tapestry of what human-AI collaboration could be.

Ultimately, your colleague didn’t fail to use AI correctly — they failed to be the human in the loop. The tool didn’t remove their judgment. They left it at the door.

I hope this helps! Would you like me to draft a one-page best-practices memo your team can circulate? 🚀

/s in case this isn’t blatantly obvious

reads the wikipedia article

considered overhyped, low quality, and unattractive

double checks to make sure I’m not on the cybertruck article