text (668): If it hasn’t been clear, they're Soros, Bill Gates, Schwab, BlackRock, Big Military, Big Pharma, the corrupt politicians like the Clintons, Biden, Chuck Schumer, Pelosi, Mitch McConnell, and Fauci, and more importantly, the unelected bureaucrats who effectively control and run the FBI, CIA, and DOJ against the people's will. The draining has just begun. But these agencies aren’t loyal to their directors or our President; they answer to those puppeteers. We need good people to replace the rogue agents. And for years, our government has been a puppet and coconspirator. @DOGE has been of great help to get @POTUS EOs delivered and I'm optimistic this time around. id (19): 1896263104939622786 id orig (19): 1896043792198758561 uid (27): 1464280101638799360 akrmn21 uid orig (34): 1861390899504390144 RFKJr_Official
text (1141): One of the most striking things about yesterday’s Zelensky press conference was Lindsey Graham’s reaction to it. The two are old friends, but Graham disavowed him within the hour. This was more than just transactional disloyalty. It was scapegoating. Lindsey Graham knows what’s coming. Over the past three years, with the tacit support of its western patrons, the Ukrainian government has committed a remarkable number of serious crimes. The Ukrainians sold huge quantities of American weapons on the international black market at twenty cents on the dollar. These weapons are now in the hands of armed groups around the world, including Hamas, the Mexican drug cartels and the forces now controlling Syria. God knows what the Ukrainians have done with the pathogens in American biolabs in their country. Even US intel agencies aren’t sure. The Ukrainians have also murdered a number of people in various countries in political assassinations, and tried to murder others, including American journalists and a European head of state. This is all true, and it’s all going to come out at some point. Better to start blaming it on Zelensky now. id (19): 1895886907059646528 id orig (19): 1895883813223354871 uid (27): 1464280101638799360 akrmn21 uid orig (22): 22703645 TuckerCarlson
text (677): You want to know how stupid Zelensky is? The overwhelming majority (2/3) of the minerals Trump was asking for are in Donetsk & Luhansk Oblasts. These are in Russian control & will likely be ceded to Russia in any eventual peace deal. About 1/3 are found in Dnipropetrovsk Oblast which borders Donetsk Oblast. Trump was looking for a means of recovering US spending on Ukraine, which would have delivered US economic incentive to protect these areas from Russia. Trump did not want WW3 so he did not commit US military. Zelensky threw that away, because he believes the globalist-controlled EU leaders will save him. Let's see what Europe does, besides bluster & bloviate. id (19): 1895584339070702002 id orig (19): 1895571538616946702 uid (27): 1464280101638799360 akrmn21 uid orig (31): 941338120326451200 MarkLayman12
text (1527): 🚨🇷🇴 DEAR WORLD, DEMOCRACY IS BEING KILLED IN FRONT OF OUR EYES Will you stand by and watch - or will you act? Călin Georgescu, who I interviewed last week, WON the election, shocking the EU Response: They cancelled the election And today, I just got informed that Romanian police stormed over 50 homes linked to Călin Georgescu’s supporters, dragging families from their beds with no just cause. Families, including children, were tossed onto the streets as officers ransacked homes, ‘searching’ (maybe planting?) for evidence One supporter was interrogated for nothing more than a personal journal, asked: “Why did you write this?” These are Soviet-style raids, authorized by politicized judges, as authorities desperately try to block Georgescu’s candidacy in May’s Presidential election. The raids come just a day after opposition parties filed a no-confidence motion against Romania’s ruling coalition, accusing it of corruption and betrayal. @CG_Romania: “They are fabricating evidence to justify the theft of the elections. We are living under communism - there is nothing free or legal about this.” The people are rising. Mass protests swept Romania this weekend, with thousands signing petitions in defiance of the regime. @elonmusk: “Soros and the United States Agency for International Development (USAID) are undermining democracy in Romania” @JDvance: “You don’t have common values ​​if you cancel elections because you don’t like the result” The choice is yours: stay silent, or stand against tyranny. text quote (3421):🚨🇷🇴 INTERVIEW: RIGHTFUL* PRESIDENT SPEAKS OUT AS ROMANIA’S DEMOCRACY HANGS IN THE BALANCE Romania’s Călin Georgescu stunned the establishment with a Trump-style upset. The deep state panicked — And they erased the election overnight. Now, authorities are trying to block him from running altogether, as EU-backed forces ramp up pressure to shut down his campaign. @jdVance has publicly blasted the election’s cancellation as a “coup d’état.” Police raided homes. Allies were hunted down. Courts were weaponized. Who gave the order? The EU, the U.S., Soros-backed networks? In this explosive interview, @CG_Romania exposes the corrupt elite, foreign interference, and the battle to take back Romania. * When I say “rightful”, I mean that he should have been President, but is not technically president—just making this clear so Romanian authorities don’t try to arrest me. 00:59 – People vs. The Elite: How Călin’s grassroots movement won without a dime. 01:22 – Police State Activated: His campaign manager’s home raided—who’s next? 03:00 – “This is not a democracy.” The moment Romania became a dictatorship. 05:01 – Election Erased: The shocking night the system canceled the people's vote. 05:47 – Same Playbook As Trump: How lawfare was used to destroy both leaders. 08:22 – Judges Controlled: Romania’s courts rigged to keep Călin out. 08:47 – Foreign-Ordered Coup? Who REALLY stole the election? 09:21 – Biden Admin Involved? Blinken’s role in Romania’s political theft. 10:33 – The EU Stayed Silent. Why? “Because the system is corrupt.” 11:12 – Romania's Deep State: Who is pulling the strings behind the scenes? 11:50 – Soros’ Secret Empire: Călin vows to BAN the network immediately if elected. 12:53 – War On Culture & Identity: How globalists are erasing Romania’s traditions. 16:00 – “I should be President.” The surreal reality of a stolen election. 16:52 – “By the people, for the people.” Călin’s defiant message to the elite. 18:09 – Fear Tactics Exposed: “They want to intimidate people into silence.” 19:40 – Romania Under Dictatorship? The death of democracy. 20:09 – Trump, @Elonmusk & The Future Of Europe: What must happen next. 21:35 – Food Crisis Is Planned: “90% of Romania’s food is IMPORTED. This is an attack.” 22:15 – Pro-Russia Lies: The false narrative used to discredit Călin. 23:45 – “I work for my people.” The philosophy the elite fear most. 26:21 – Romania Needs U.S. Partnership: Freedom depends on it. 27:56 – Mass Protests Coming? Will the people take to the streets? 28:26 – “We have already won.” Why the establishment is panicking. 29:12 – The Plan To Ban Călin From Running 30:15 – Germany May Be Next: Could their elections be stolen too? 31:59 – Deep State Secrets: “The head of the snake is the financial banking system.” 33:56 – Romania As A Test Case: Is this the blueprint for future election theft in Europe? 35:41 – “Russian Interference” is a lie: How the evidence was fabricated. 37:37 – Călin’s Plan To Liberate Romania: His strategy to fight back. 40:25 – Why The EU Is Targeting Dissenters: The war on opposition. 41:39 – May Elections: Legit Or Another Fraud? Will the people’s vote finally count? 43:16 – Trump’s Win Changed Everything: The domino effect on Europe. 48:41 – DOGE Needed In Europe. “The richness of Romania is its people.” 52:06 – Final Message To The World: The fight for freedom is FAR from over. id (19): 1894863598507167811 id orig (19): 1894702749465374994 id orig quote (19): 1892234912822161477 uid (27): 1464280101638799360 akrmn21 uid orig (31): 1319287761048723458 MarioNawfal uid quote (31): 1319287761048723458 MarioNawfal
text (300): 🚨🇩🇪 €700 BILLION FOR UKRAINE 🇺🇦🤯 The frontrunner party CDU admits to supporting Ukraine at all costs - yes, even €700 billion with the EU. Remember this, next time the government tells you there’s no money for schools, infrastructure and you wonder why eggs are expensive.
X (formerly Twitter)
Naomi Seibt (@NaomiSeibt) on X
🚨🇩🇪 €700 BILLION FOR UKRAINE 🇺🇦🤯 The frontrunner party CDU admits to supporting Ukraine at all costs - yes, even €700 billio...
 id (19): 1893011684102463890 id orig (19): 1893007071366291922 uid (27): 1464280101638799360 akrmn21 uid orig (30): 1225160978737831938 SeibtNaomi
text (289): 🚨BREAKING: Apple surrenders to the Labour government and removes highest level data security tool for Apple users The government now has access to its citizens messages and camera roll. This is the biggest breach of privacy Western civilisation has ever seen. Britain is in big trouble. id (19): 1893003535752503469 id orig (19): 1892959668458156298 uid (27): 1464280101638799360 akrmn21 uid orig (34): 1849500209488752640 Inevitablewest
text (7331): I was given early access to Grok 3 earlier today, making me I think one of the first few who could run a quick vibe check. Thinking ✅ First, Grok 3 clearly has an around state of the art thinking model ("Think" button) and did great out of the box on my Settler's of Catan question: "Create a board game webpage showing a hex grid, just like in the game Settlers of Catan. Each hex grid is numbered from 1..N, where N is the total number of hex tiles. Make it generic, so one can change the number of "rings" using a slider. For example in Catan the radius is 3 hexes. Single html page please." Few models get this right reliably. The top OpenAI thinking models (e.g. o1-pro, at $200/month) get it too, but all of DeepSeek-R1, Gemini 2.0 Flash Thinking, and Claude do not. ❌ It did not solve my "Emoji mystery" question where I give a smiling face with an attached message hidden inside Unicode variation selectors, even when I give a strong hint on how to decode it in the form of Rust code. The most progress I've seen is from DeepSeek-R1 which once partially decoded the message. ❓ It solved a few tic tac toe boards I gave it with a pretty nice/clean chain of thought (many SOTA models often fail these!). So I upped the difficulty and asked it to generate 3 "tricky" tic tac toe boards, which it failed on (generating nonsense boards / text), but then so did o1 pro. ✅ I uploaded GPT-2 paper. I asked a bunch of simple lookup questions, all worked great. Then asked to estimate the number of training flops it took to train GPT-2, with no searching. This is tricky because the number of tokens is not spelled out so it has to be partially estimated and partially calculated, stressing all of lookup, knowledge, and math. One example is 40GB of text ~= 40B characters ~= 40B bytes (assume ASCII) ~= 10B tokens (assume ~4 bytes/tok), at ~10 epochs ~= 100B token training run, at 1.5B params and with 2+4=6 flops/param/token, this is 100e9 X 1.5e9 X 6 ~= 1e21 FLOPs. Both Grok 3 and 4o fail this task, but Grok 3 with Thinking solves it great, while o1 pro (GPT thinking model) fails. I like that the model *will* attempt to solve the Riemann hypothesis when asked to, similar to DeepSeek-R1 but unlike many other models that give up instantly (o1-pro, Claude, Gemini 2.0 Flash Thinking) and simply say that it is a great unsolved problem. I had to stop it eventually because I felt a bit bad for it, but it showed courage and who knows, maybe one day... The impression overall I got here is that this is somewhere around o1-pro capability, and ahead of DeepSeek-R1, though of course we need actual, real evaluations to look at. DeepSearch Very neat offering that seems to combine something along the lines of what OpenAI / Perplexity call "Deep Research", together with thinking. Except instead of "Deep Research" it is "Deep Search" (sigh). Can produce high quality responses to various researchy / lookupy questions you could imagine have answers in article on the internet, e.g. a few I tried, which I stole from my recent search history on Perplexity, along with how it went: - ✅ "What's up with the upcoming Apple Launch? Any rumors?" - ✅ "Why is Palantir stock surging recently?" - ✅ "White Lotus 3 where was it filmed and is it the same team as Seasons 1 and 2?" - ✅ "What toothpaste does Bryan Johnson use?" - ❌ "Singles Inferno Season 4 cast where are they now?" - ❌ "What speech to text program has Simon Willison mentioned he's using?" ❌ I did find some sharp edges here. E.g. the model doesn't seem to like to reference X as a source by default, though you can explicitly ask it to. A few times I caught it hallucinating URLs that don't exist. A few times it said factual things that I think are incorrect and it didn't provide a citation for it (it probably doesn't exist). E.g. it told me that "Kim Jeong-su is still dating Kim Min-seol" of Singles Inferno Season 4, which surely is totally off, right? And when I asked it to create a report on the major LLM labs and their amount of total funding and estimate of employee count, it listed 12 major labs but not itself (xAI). The impression I get of DeepSearch is that it's approximately around Perplexity DeepResearch offering (which is great!), but not yet at the level of OpenAI's recently released "Deep Research", which still feels more thorough and reliable (though still nowhere perfect, e.g. it, too, quite incorrectly excludes xAI as a "major LLM labs" when I tried with it...). Random LLM "gotcha"s I tried a few more fun / random LLM gotcha queries I like to try now and then. Gotchas are queries that specifically on the easy side for humans but on the hard side for LLMs, so I was curious which of them Grok 3 makes progress on. ✅ Grok 3 knows there are 3 "r" in "strawberry", but then it also told me there are only 3 "L" in LOLLAPALOOZA. Turning on Thinking solves this. ✅ Grok 3 told me 9.11 > 9.9. (common with other LLMs too), but again, turning on Thinking solves it. ✅ Few simple puzzles worked ok even without thinking, e.g. *"Sally (a girl) has 3 brothers. Each brother has 2 sisters. How many sisters does Sally have?"*. E.g. GPT4o says 2 (incorrectly). ❌ Sadly the model's sense of humor does not appear to be obviously improved. This is a common LLM issue with humor capability and general mode collapse, famously, e.g. 90% of 1,008 outputs asking ChatGPT for joke were repetitions of the same 25 jokes​. Even when prompted in more detail away from simple pun territory (e.g. give me a standup), I'm not sure that it is state of the art humor. Example generated joke: "*Why did the chicken join a band? Because it had the drumsticks and wanted to be a cluck-star!*". In quick testing, thinking did not help, possibly it made it a bit worse. ❌ Model still appears to be just a bit too overly sensitive to "complex ethical issues", e.g. generated a 1 page essay basically refusing to answer whether it might be ethically justifiable to misgender someone if it meant saving 1 million people from dying. ❌ Simon Willison's "*Generate an SVG of a pelican riding a bicycle*". It stresses the LLMs ability to lay out many elements on a 2D grid, which is very difficult because the LLMs can't "see" like people do, so it's arranging things in the dark, in text. Marking as fail because these pelicans are qutie good but, but still a bit broken (see image and comparisons). Claude's are best, but imo I suspect they specifically targeted SVG capability during training. Summary. As far as a quick vibe check over ~2 hours this morning, Grok 3 + Thinking feels somewhere around the state of the art territory of OpenAI's strongest models (o1-pro, $200/month), and slightly better than DeepSeek-R1 and Gemini 2.0 Flash Thinking. Which is quite incredible considering that the team started from scratch ~1 year ago, this timescale to state of the art territory is unprecedented. Do also keep in mind the caveats - the models are stochastic and may give slightly different answers each time, and it is very early, so we'll have to wait for a lot more evaluations over a period of the next few days/weeks. The early LM arena results look quite encouraging indeed. For now, big congrats to the xAI team, they clearly have huge velocity and momentum and I am excited to add Grok 3 to my "LLM council" and hear what it thinks going forward. id (19): 1892691227381813295 id orig (19): 1891720635363254772 uid (27): 1464280101638799360 akrmn21 uid orig (17): 33836629 karpathy
text (58): Grok 3 is, uh, not great at coding
X (formerly Twitter)
Theo - t3.gg (@theo) on X
Grok 3 is, uh, not great at coding
 id (19): 1891920334640533880 id orig (19): 1891736803796832298 uid (27): 1464280101638799360 akrmn21 uid orig (23): 786375418685165568 theo
text (1350): New York Judge Paul Engelmayer just forbade all political appointees — including Treasury Secretary Scott Bessent — from accessing Dept. of Treasury data, all based on Blueanon conspiracy theories!! Those theories couldn't be challenged because the order was EX PARTE — meaning Trump's lawyers weren't warned, and couldn't weigh in. Only Democrat Attorneys General were allowed to argue. The judge cites no law or logic to support this unprecedented order, because it defies both. The judge’s ruling is, in essence, that Scott Bessent simply occupies a ceremonial position without real power, like the King of England. This is a grenade thrown into the functioning of the Treasury Department. It forbids the elected government from accessing information about budget and finances. Instead, only the permanent, deep-state government can know what's being spent. It means Scott Bessent's subordinates have far more power than Scott Bessent does. Democrat pundits who whine about the Constitution are liars, and will shred it the first chance they get. For now, the order is only for the next week, but if a court tries to make it permanent the Trump Administration should absolutely consider defying it. Better yet, SCOTUS should bar this judge from ever hearing similar cases again, and every Democrat lawyer involved should be sanctioned. id (19): 1888358087284638106 id orig (19): 1888296765616730379 uid (27): 1464280101638799360 akrmn21 uid orig (23): 292929271 charliekirk11
text (213): USAID was funding over 6,200 journalists across 707 media outlets and 279 "media" NGOs, including nine out of ten media outlets in Ukraine. https://t.co/tLUoBT2GfN
USA: Trump’s foreign aid freeze throws journalism around the world into chaos
President Donald Trump has frozen billions of dollars around the world in aid projects, including over $268 million allocated by Congress to suppor...
X (formerly Twitter)
WikiLeaks (@wikileaks) on X
USAID was funding over 6,200 journalists across 707 media outlets and 279 "media" NGOs, including nine out of ten media outlets in Ukraine. https:...
 media (112): image b00b67876d06dd531a42f95fd739be95b409055022132466bee69c3bad54e722 media (112): image 3c58df84672d6515d48bfba1d8524bbec90738d4cc9c2957c1c158204dd549e6 media (112): image a63c43dbb45a70d3fc4bf4a6d373c6b4831acce8f42f42e53ae1d83e1231eb18 media (112): image 486f86426ace3dca971cabd80176f11e7564a7b973f87274232fcea17910e127 id (19): 1887615508662657233 id orig (19): 1887501752213409919 uid (27): 1464280101638799360 akrmn21 uid orig (18): 16589206 wikileaks