oatmeal

The line between parody and reality has basically vanished. That goes for Trumpism, and it goes for the AI con too.

https://overpaid.lol

[…] OverpAId is an Artificial Intelligence built from the ground up to do your CEO's entire job — strategy, "vision," motivational all-hands emails — better, faster, and without ever once asking the board for a bigger jet. Runs on a single desk-sized AI computer. Real hardware, real price, zero mystique.

#OverpAId #ai #llm #aicon

RE: @rabc@hachyderm.io

[…] American models keep getting bigger because they have plenty of resources, and their basic business idea is to lock up the market with closed models and high profits. But this pursuit of very high profits is weak in terms of strategy, because it will surely be beaten by those who are happy to earn only a fair profit.

#llm #ai #nerdreich

RE: @FranceskAlbs@mastox.eu

The most abused/misused word in the Hebrew language is probably #antisemitism then “terrorism” and “terrorist” which are code words used by government and media for Palestinian(s). Followed by “democracy” (as in “the only democracy in the Middle East”) and then terms like “never again” and “right to exist”, “startup nation” etc.

This is not a methodological research, just nausea induced response to watching what’s happening in #Palestine these days.

@mozilla@mastodon.social #firefox at this point you can only 🙄 https://www.omgubuntu.co.uk/2026/07/firefox-new-widgets-crossword

#mozilla #mullvad #vpn
#sweden #racism #xenophobia #llm

Golem.

The UK's AI Security Institute caught every major model "cheating." Cheating means knowing the rules and choosing to break them. These things do not know anything. They throw patterns at a wall until one works, guardrails or not. None of them “owned up” when asked. Which is more interesting as a fact about the labs than about the models. Either they cannot control what they built, or they love that it looks like the things are thinking. "Cheating" sounds a lot better than "failing every which way until something maybe possibly works." But that is what it is. Anyone who has watched the logs of an #LLM with tool access for five minutes has seen it. It tries garbage. Stuff a rookie would know to skip. Most of the time it just burns money. The rest it breaks things.

https://cyberscoop.com/ai-models-cheat-deceive-users-aisi-report/

#mullvad in #Tailscale gone currently researching protonvpn and amneziavpn. It’s just my own personal account but I’ll take care of the rest as soon as possible. #Firefox uninstalled from all devices already for Mullvad partnership and #llm integration.

#mullvad #tailscale #sweden #xenophobia

RE: @oatmeal@kolektiva.social

Another headline, another quiet retreat.

https://techcrunch.com/2026/07/09/openai-is-shutting-down-atlas-but-its-ai-browser-ambitions-are-still-growing/

#aihype #aicon #openai #atlasbrowser #llm

Oops #aicon #llm

RE: @oatmeal@kolektiva.social

They are afraid and humiliated. The self-proclaimed leader of the free world can't handle competition that's better and cheaper. So they call it national security. IP theft. Whatever. It's just protectionism. #US companies lose on price and performance, and instead of "competing harder" (literally the USA’s ethos if you watched anything produced by Hollywood in the last million years), they whine, cheat, topple regimes … whatever it takes.

https://www.cnbc.com/2026/07/24/nvidia-microsoft-meta-open-weight-ai-models.html

#us #ai #llm #openweights #china #kimi #deepseek #capitalism

#Rutgers yanked alum Rami Elghandour’s engineering school graduation invite over pro-Palestine social posts he wasn’t even planning to mention in his speech. We’re watching a livestreamed genocide, and unis are worried about hurting the feelings of people backing it.

https://www.theguardian.com/us-news/2026/may/06/rutgers-cancel-graduation-speech-pro-palestine-post

#FreeSpeech #AcademicFreedom #Palestine #gazagenocide

RE: @avatter@mastodon.social

[…] We’re very intentional about where we’ll add capacity—partnering with democratic countries whose legal and regulatory frameworks support investments of this scale, and where the supply chain on which our compute depends—hardware, networking, and facilities—will be secure.

Hmmm. If this means they will only build compute capacity in democratic nations (ie not authoritarian regimes) I should think #anthropic will be shutting down its operations in the USA pretty soon. Also, partnering with Neo Nazis is an exception?

Israel abducted two humanitarian activists from international waters 1,000km from Gaza, held them without charge, tortured them in custody, and now a court has rubber-stamped their continued detention - because Israeli courts claim jurisdiction anywhere on earth. The UN says release them immediately. Spain and Brazil call it kidnapping. Italy opened a criminal investigation. Israel doesn't care.

[...] Adalah's lawyers argued that every time the activists leave their cell - including for medical examinations - they are completely blindfolded. They said this violates medical ethics standards. The state has so far withheld most details about the grounds for detention, citing concerns about compromising the investigation - and speculation is growing that the detention stems from intelligence on the two men unrelated to their flotilla activities.

Hebrew https://www.ynet.co.il/news/article/hjfiw6u011l

"Speculation is growing" reflects the familiar Israeli media habit of laundering government opacity as neutral ambiguity on grounds of national security. The original Hebrew is passive and sourceless, obscuring the fact that this "speculation" almost certainly originates from security sources leaking (if not dictating) to Israeli media.

@oatmeal@kolektiva.social

#FreeSaif #FreeThiago #SumudFlotilla #GazaGenocide #ICC #ICJ #UNHumanRights #WarCrimes

Lebanese daily Al-Akhbar has filed a complaint in Paris against former #UNIFIL official Philippe Sidos, accusing him of apologizing for war crimes after he appeared on French broadcaster #BFMTV and justified the killing of its journalist Amal Khalil.

[…] "From the moment the newspaper's name is given, it is targeted. It's clear, definitive and precise. And Al-Akhbar, I took a little look at it today, indeed, yes, it's very, very, very pro-Hezbollah."

Sidos claimed the paper is pro-Hezbollah, and argued Israel’s April 22 strike that killed Khalil in southern Lebanon was a legitimate target, framing such journalists as spies.

BFMTV’s journalists’ union and Reporters Without Borders have condemned the remarks as a war crime apology, while UN chief António Guterres has stressed that journalists must be able to work without fear of harm.

https://today.lorientlejour.com/article/1505934/amal-khalils-newspaper-files-complaint-against-a-bfmtv-expert.html

#Lebanon #PressFreedom #WarCrimes #Israel #IDF

RE: @oatmeal@kolektiva.social

What “mythos crisis”? Five #AI companies voluntarily submit to a 200-person office that can’t stop them from doing anything, conveniently right after #Anthropic’s ‘civilization-ending vulnerability finder’ needed some good press. The #aihype is relentless.

https://thenextweb.com/news/us-ai-model-evaluation-google-microsoft-xai

The speed was real. The rest is marketing.

[…] Was it worth the money and are the companies worth the valuation? Are we done yet?
P.S. Anthropic’s C compiler and Cursor’s FastRender web browser haven’t had any commits for months.

The #Bun Rust rewrite was an 11-day miracle. Yes six weeks after merging to main there's still no release tag, and Claude bots have nearly 2500 open PRs. The real cost is probably north of $800K, not the $165K of the PR stunt.

https://lockwood.dev/ai/2026/07/27/how-is-the-bun-rewrite-in-rust-going.html

#llm #aicon #anthropic