cegees

Our own experience makes us stubborn old people when it comes to the future.

2nd place in the Gemini Croquettes Contest.

[cegees @cegees]

Ups.

[cegees @cegees]

@MausethSeth @timruss2 That's why they're considering German or South Korean submarines.

[Liz Webster @LizWebsterSBF]

Mark Zuckerberg settles an $8bn lawsuit to avoid testifying under oath over Cambridge Analytica and walks away without public accountability.

This is surveillance capitalism in action: law-breaking treated as a cost of doing business, executives shielded, power untouched.

This is exactly why strong regulation matters and why weakening it, as the US model does, hands democracy to tech oligarchs, not voters.

[Joel Willans @VFinnishProbs]

Says the son of an immigrant, married to an immigrant, leading a nation built by immigrants.

[Vitalii «Donetsk» Ovcharenko 🇺🇦 @Vit_Ovchar]

The photo shows the Ukrainian 🇺🇦 army in Iraq 🇮🇶 helping the United States 🇺🇸 fight against the regime of Saddam #Hussein.

When the United States 🇺🇸 asked for help, Ukraine 🇺🇦 responded by sending its troops and joining the U.S. 🇺🇸 coalition.
In Iraq 🇮🇶, 18 Ukrainian soldiers were killed while serving alongside the United States 🇺🇸.

Today, Ukrainians are not asking the United States to fight for us.
We are simply asking you not to side with the aggressor - Russia 🇷🇺.

[Euromaidan Press @EuromaidanPress]

The US Treasury quietly removed sanctions from several companies previously accused of supplying Russia's military - offering no explanation for the move.

The delisted firms from Cyprus, Dubai, Turkey, and Finland had been linked to shipping military equipment, machine tools, and electronic components to Russia.

The Russian entities these companies supplied remain under sanctions. The suppliers themselves are now free to operate.

US authorities did not indicate whether the removals reflect changes in compliance, enforcement priorities, or broader sanctions policy.

🔗Read more: https://t.co/TOMwmcf6QZ

[Kerry Kennedy @KerryKennedyRFK]

Three years and one month from today, I’m going to grab a pickax and pull those letters off that building, but I’m going to need help holding the ladder. Are you in? Applying for my carpenter’s card today, so it’ll be a union job!!!

[Tim Vollert @Tim_Vollert]

Ich denke nicht dass wir das ignorieren sollten. Eine der größten Innovationen Europas sind EUV-Maschinen von ASML.
Chinesisch-stämmige ASML Ingenieure sind zurück nach China und haben in Shenzen einen Prototyp der Maschine nachgebaut. Zuvor gab es schon Industriespionagefälle.

[NikTek @NikTek]

YouTube auto-dub feature is probably the worst AI implementation on this platform I've ever seen.

I have skipped a ton of videos because I thought the creator was using some cheap AI voice-over.

Turns out, most of the time the AI detected that the original track is of another language and translated it to English, in this case here it says that the original track is Korean (when in fact it was English) and it just turned on auto-dub for the viewer automatically.

If you're a YouTube creator I strongly suggest you take a look at auto-dub on your videos because you might be loosing viewers if the AI thinks that your original track is another language.

[Alex Veremeyenko @alex_verem]

This paper from Harvard and MIT quietly answers the most important AI question nobody benchmarks properly:

Can LLMs actually discover science, or are they just good at talking about it?

The paper is called “Evaluating Large Language Models in Scientific Discovery”, and instead of asking models trivia questions, it tests something much harder:

Can models form hypotheses, design experiments, interpret results, and update beliefs like real scientists?

Here’s what the authors did differently 👇

• They evaluate LLMs across the full discovery loop hypothesis → experiment → observation → revision
• Tasks span biology, chemistry, and physics, not toy puzzles
• Models must work with incomplete data, noisy results, and false leads
• Success is measured by scientific progress, not fluency or confidence

What they found is sobering.

LLMs are decent at suggesting hypotheses, but brittle at everything that follows.

✓ They overfit to surface patterns
✓ They struggle to abandon bad hypotheses even when evidence contradicts them
✓ They confuse correlation for causation
✓ They hallucinate explanations when experiments fail
✓ They optimize for plausibility, not truth

Most striking result:

`High benchmark scores do not correlate with scientific discovery ability.`

Some top models that dominate standard reasoning tests completely fail when forced to run iterative experiments and update theories.

Why this matters:

Real science is not one-shot reasoning.

It’s feedback, failure, revision, and restraint.

LLMs today:

• Talk like scientists
• Write like scientists
• But don’t think like scientists yet

The paper’s core takeaway:

Scientific intelligence is not language intelligence.

It requires memory, hypothesis tracking, causal reasoning, and the ability to say “I was wrong.”

Until models can reliably do that, claims about “AI scientists” are mostly premature.

This paper doesn’t hype AI. It defines the gap we still need to close.

And that’s exactly why it’s important.

[Aaron Reichlin-Melnick @ReichlinMelnick]

This is a BIG deal for a lot of Americans. When Trump's new travel restriction kicks in, if you are an American and you marry someone from Cuba, Venezuela, Nigeria, Haiti, or any of 35 other countries, you will be banned from having your husband or wife join you in this country.

[JMireles @JMireles7]

@StephenMoore @realDonaldTrump 😂😂….So the CPI just started ignoring all other data points now. They’re literally using 3 of the 16 data markers to get this number. Aka, cooking the books and not even trying to hide it…

[George W. Bush Presidential Center @TheBushCenter]

“Putin is the problem; Ukraine is not. Putin is the obstacle to peace; Ukraine is not. Putin is the one who should feel as much pressure as possible, not Ukraine. Putin is the one who started this war; Ukraine didn’t,” says David J. Kramer, executive director of the Bush Institute.

Kramer suggests that the United States, along with European allies, should increase military assistance for Ukraine and ramp up sanctions against Russia and its accomplices. “Ukrainians want to end the war as soon as possible, but they also reject, and rightly so, the idea of territorial concessions,” says Kramer.

Read more on how tightening sanctions can help end the war in Ukraine that Putin started: https://t.co/DbwUEu5QKe

[Rep. Joyce Beatty @RepBeatty]

For the record. This was not unanimous. I was muted on the call and not allowed to speak or voice my opposition to this  move. Also for the record, this was not on the agenda. This was not consensus. This is censorship.

[Pete @splendid_pete]

So much bullshit. Fewer working hours are not bad work ethics. Europeans work fewer hours, not less effectively. Output per hour in Germany, France, and the Netherlands is at or above the US. Productivity is about what you produce per hour, not how long you sit at work.

EU government spending is roughly 49 to 51 percent of GDP, and that covers healthcare, pensions, education, childcare, disability, infrastructure, and public investment. The US sits around 36 to 38 percent, but Americans then pay privately for the same things through insurance premiums, medical bills, student loans, and fees that never show up as taxes. GDP does not care whether you pay a hospital with taxes or a credit card.

Government spending share tells you nothing about efficiency, productivity, or living standards. The US only looks smaller because healthcare, pensions, and education are privatized and show up later as household debt and insurance premiums instead of taxes. Same cost, worse distribution.

Dalio’s actual concern about Europe is demographics and capital markets depth, not that public spending makes GDP fake or people lazy. Quoting him like that is just misleading.

[Ben Phillips @benphillips76]

This is beautiful. The smiles! Bondi beach hero Ahmed Al-Ahmed with the doctors looking after him at the hospital in Sydney.

And…

The doctors are *also* Syrian!

Tamer Al-Kahil from Homs.
Anas Natfaji from Aleppo.

*Three* heroes.

[ᗰᗩƳᖇᗩ @LePapillonBlu2]

President Joe Biden rang the bell today at the end of chemo. 🙌🏽

[Michael Weiss @michaeldweiss]

The New York and North Carolina Young Republicans helped organize this tour for ROC on the Hill. The New York Young Republican Club just hosted AfD deputy Markus Fronhmaier, described in a strategy document shared with the Russian Presidential Administration as someone "under our absolute control." There are no guardrails anymore. https://t.co/kTUpIDRxR6

[cegees @cegees]

@PeteHegseth @POTUS A blueprint for China to take Taiwan and for Russia to justify taking over Ukraine. Brave New World.

[I Smoked A Predilection For Nipples @BlackKnight10k]

Here is Trump's White House physician handing out fentanyl like candy during his first administration. His nickname in the WH was literally "the candyman."

Lots of ketamine in there too.

[@ChrisStoecker @ChrisStoecker]

Jetzt schwarz auf weiß: Das Wirtschaftsministerium von Katherina Reiche bestellt einen Expertenbericht (schon mit nahegelegtem Spin), bekommt ihn - und schreibt ihn dann vor Veröffentlichung so um, dass das darin steht, was Reiche will. "mehr EE nötig"(l.) = gestrichen.